• Chinese imports entering through Wal-Mart in 2013 likely totaled at least $49.1 billion and the combined effect of imports from and exports to China conducted through Wal-Mart likely accounted for 15.3 percent of the growth of the total U.S. goods trade deficit with China between 2001 and 2013.
• The Wal-Mart-based trade deficit with China alone eliminated or displaced over 400,000 U.S. jobs between 2001 and 2013.
• The manufacturing sector and its workers have been hardest hit by the growth of Wal-Mart’s imports. Wal-Mart’s increased trade deficit with China between 2001 and 2013 eliminated 314,500 manufacturing jobs, 75.7 percent of the jobs lost from Wal-Mart’s trade deficit. These job losses are particularly destructive because jobs in the manufacturing sector pay higher wages and provide better benefits than most other industries, especially for workers with less than a college education.
• Wal-Mart has announced plans to create opportunities for American manufacturing by “investing in American jobs.” To date, very few actual U.S. jobs have been created by this program, and since 2001, the growing Wal-Mart trade deficit with China has displaced more than 100 U.S. jobs for every actual or promised job created through this program.

China has achieved its rapidly growing trade surpluses by manipulating its currency: it invests hundreds of billions of dollars per year in U.S. Treasury bills, other government securities, and private foreign assets to bid up the value of the dollar and other currencies and thereby lower the cost of its exports to the United States and other countries. China has also repressed the labor rights of its workers and suppressed their wages, making its products artificially cheap and further subsidizing its exports. Wal-Mart has aided China’s abuse of labor rights and its violations of internationally recognized norms of fair trade by providing a vast and ever-expanding conduit for the distribution of artificially cheap and subsidized Chinese exports to the United States.

China trade and U.S. job loss

Exports support jobs in the United States, and imports displace them. Thus, the net effect of trade flows on employment must be based on an analysis of the trade balance. This Briefing Paper calculates the employment effects of growing goods trade deficits by using an input-output model that estimates the direct and indirect labor requirements of producing output in a given domestic industry. The model includes 195 U.S. industries, 77 of which are in the manufacturing sector.1

The model estimates the labor that would be required to produce a given volume of exports, and the labor that is displaced when a given volume of imports is substituted for domestic output.2 The job losses presented here represent an estimate of what total employment levels would have been in the absence of growing trade deficits.3

U.S. exports to China in 2001 supported 161,400 jobs, but U.S. imports displaced production that would have supported 1,127,700 jobs, as shown in the bottom half of Table 1. Therefore, the $84.1 billion goods trade deficit in 2001 displaced nearly 1 million jobs in that year. Net job displacement rose to 4,123,400 in 2013. Growth in trade deficits with China has reduced demand for goods produced in every region of the United States and has led to job displacement in all 50 states and the District of Columbia. The overall China trade and job loss estimates in this report are based on the findings reported in Kimball and Scott (2014).

Table 1

Wal-Mart’s role

Given its enormous size and the fact that it sells manufactured goods, which have been the primary Chinese export to the U.S. in recent years, it is natural to try to estimate the role of Wal-Mart as a conduit for Chinese trade. We find that a conservative estimate is that Wal-Mart accounted for approximately 11.2 percent of total U.S. goods imports from China between 2001 and 2013. This estimate is based on published reports on Wal-Mart trade with China between 2001 and 2004, including Wal-Mart’s own estimates of its imports from China, on more recent published data on ocean trade (by company), and on the relationship between total Wal-Mart sales in the United States and personal consumption expenditures on goods from the GDP accounts (BEA 2015).

Wal-Mart provided its own estimate for the value of imports from China in its fiscal year ending January 31, 2004 (Wal-Mart 2007). Most of these goods were imported in 2003, and the Wal-Mart share of total imports from China in that year was 11.9 percent. Bianco and Zellner (2003) and Bianco (2006) have also attempted to construct estimates of Wal-Mart’s imports from China and have reported imports that yield shares that are similar to Wal-Mart’s own estimates, with the lowest share reported as 11.2 percent in 2004. Since 2007, evidence strongly suggests that this share has not shrunk (and may have risen). For example, The Journal of Commerce produces annual reports of total U.S. imports and exports of goods via ocean container transport.4 While this is a partial and incomplete accounting, it does show that Wal-Mart was the top U.S. importer of ocean container freight in every year between 2001 and 2013, and its share of top 100 imports remained stable in a range from 12.1 percent to 14.8 percent of total imports of the top 100 importers.5

Limited data on total imports by company are also available from shipments data collected by the U.S. Customs and Border Protection agency.6 Data on Wal-Mart imports are available for only two comparable months in the study period: November 2007 and 2012. The available information reports total imports in both kilograms and container equivalents (twenty-foot equivalent units or TEUs). The Wal-Mart share of total imports from China increased in both kilograms and TEUs in this period (Panjiva.com 2015). In short, the 2003 share of imports accounted for by Wal-Mart as estimated by the company itself (11.2 percent) has likely only grown since then. However, for this report we make the conservative assumption that it has remained stable.

But a stable share of Wal-Mart imports implies rapid growth in volumes. U.S. goods imports from China increased $336.1 billion between 2001 and 2013, as shown in the top half of Table 1, an increase of 329 percent. If Wal-Mart’s share of U.S. imports from China remained stable in this period at 11.2 percent, this implies that its imports increased from $11.4 billion in 2001 to $49.1 billion in 2013, an increase of $37.6 billion. As it is a retailer and not a manufacturer, Wal-Mart likely exports only a negligible amount to China. Our best estimate is that Wal-Mart accounts (at most) for roughly 1.0 percent of total U.S. exports to China.7 This in turn implies that Wal-Mart was responsible for a $36.7 billion increase in the U.S. trade deficit with China between 2001 and 2013.

The Wal-Mart trade deficit displaced 125,800 jobs in 2001 and 541,300 jobs in 2013.8 Thus, Wal-Mart was responsible for displacing at least an additional 415,400 U.S. jobs between 2001 and 2013, as shown in the bottom half of Table 1 and in Figure A. While Wal-Mart was responsible for 11.2 percent of U.S. imports in this period, it was responsible for 13.2 percent of the U.S. job losses due to growing trade deficits with China (Table 1). Since Wal-Mart’s exports to China were negligible, the rapid growth of its imports had a proportionately bigger impact on the U.S. trade deficit and job losses than overall U.S. trade flows with China (since the rest of U.S. trade with China does include significant U.S. exports to that country). On average, each of the 4,835 stores Wal-Mart operated in the United States in fiscal 2014 (Wal-Mart Stores Inc. 2014) was responsible for the loss of about 86 U.S. jobs due to the growth of Wal-Mart’s trade deficit with China between 2001 and 2013.

Figure A

These job loss estimates are conservative because goods sold at Wal-Mart are primarily durable and nondurable consumer goods, such as furniture, apparel and textiles, toys, and sporting goods. These are particularly labor-intensive manufacturing industries and support more jobs per $1 billion of imports than more capital-intensive goods such as machine tools, motor vehicles and parts, and aircraft and parts imported by other U.S. firms.

Job losses in manufacturing account for 75.7 percent of total jobs displaced due to the growing U.S. trade deficit with China in this period (Kimball and Scott 2014, Table 3). Jobs in the manufacturing sector pay higher wages and provide better benefits than most other industries, especially for workers with less than a college education. Manufacturing also employs a greater share of such workers than other sectors (Scott 2013).

The job displacement estimates in this study are conservative. They include only the jobs directly or indirectly displaced by trade, and exclude jobs in domestic wholesale and retail trade or advertising; they also exclude re-spending employment.9 They also do not account for the fact that during the Great Recession of 2007–2009, and continuing through 2013, jobs displaced by China trade reduced wages and spending, which led to further job losses.

Further, the labor-market effects of the U.S. trade deficit with China are not limited to job loss and displacement and the associated direct wage losses. Competition with low-wage workers from less-developed countries such as China has driven down wages for workers in U.S. manufacturing and reduced the wages and bargaining power of similar, non-college-educated workers throughout the economy, as previous EPI research has shown (Bivens 2013). The affected population includes essentially all workers with less than a four-year college degree—such workers make up roughly 70 percent of the workforce, or about 100 million workers (U.S. Census Bureau 2015).

The workers affected by this job displacement include millions whose jobs were not lost but whose wages were held down because of increased labor market competition with the job losers. As earlier EPI research has shown, trade with China between 2001 and 2011 displaced 2.7 million workers, who suffered a direct loss of $37.0 billion in reduced wages alone when re-employed in non-traded industries in 2011 (Scott 2013). In addition, the nation’s 100 million non-college educated workers suffered a total loss of roughly $180 billion due to increased trade with low-wage countries. These indirect wage losses were nearly five times greater than the direct losses suffered by workers displaced by China trade, and the pool of affected workers was nearly 40 times larger (100 million non-college-educated workers versus 2.7 million displaced workers).10

Wal-Mart’s U.S. manufacturing promises

In 2013 Wal-Mart announced a plan to purchase “$250 billion in products that support the creation of American jobs” by 2023 by increasing purchases of U.S. manufactured goods (Loeb 2013, Wal-Mart 2015a).11 To date, very few actual U.S. manufacturing jobs have been created as a result of this commitment. Wal-Mart remains, by far, the top importer of ocean shipping containers in the United States with total imports of more than 775,000 container-equivalents (TEUs) in 2014, exceeding total imports by Target, the number two importer, by more than 250,000 TEUs (48.7 percent, more than total Target imports) (Journal of Commerce 2015). In addition, about two-thirds of what Wal-Mart calls American-made goods are actually groceries, which support few U.S. manufacturing jobs (Alliance for American Manufacturing 2015).

In 2015, Wal-Mart’s publicly available list of manufacturing jobs that have been or will be created in the United States includes fewer than 4,100 specific U.S. manufacturing jobs, and many of those are promised jobs that firms “will create” up to 10 years in the future (Wal-Mart 2015c). Since 2001, Wal-Mart’s growing trade deficit with China has displaced more than 100 U.S. jobs for every job that Wal-Mart has created in the United States through its “Invest in American Jobs” program.” Meanwhile, the U.S. goods trade deficit with China increased by $23.9 billion (7.5 percent) in 2014 (Scott 2015). Continuing growth in that trade deficit and in Wal-Mart imports will likely displace many times more manufacturing jobs than Wal-Mart creates in the United States over the next decade.

Conclusion

The growing goods trade deficit with China displaced 3.2 million U.S. jobs in the United States between 2001 and 2013, and it has been a prime contributor to the crisis in manufacturing employment over the past 15 years. Due to its own growing trade deficit with China, Wal-Mart alone was responsible for the loss of more than 400,000 U.S. jobs, 13.2 percent of total U.S. jobs lost in this period. The current unbalanced U.S.-China trade relationship is bad for both countries, and Wal-Mart has played a major role in creating that imbalance. The United States is piling up foreign debt, losing export capacity, and facing a more fragile macroeconomic environment.

Meanwhile, China has become dependent on the U.S. consumer market for employment generation, has suppressed the purchasing power of its own middle class with a weak currency, and, most importantly, has purchased trillions of dollars of hard-currency reserves in low-yielding, government securities and other financial assets, instead of investing these funds in public goods that could benefit Chinese consumers and workers. In order to artificially and illegally hold down the value of its currency, and thereby lower the cost of its exports to the United States and other countries, China has purchased nearly $5 trillion in U.S. Treasury bills and other government securities and private assets (IMF 2015, SWFI 2015) since it entered the WTO in 2001. It has also repressed the labor rights and wages of its workers, making its exports artificially cheap, further subsidizing its exports. Wal-Mart has aided China’s abuse of labor rights and its violations of internationally recognized norms of fair trade behavior by providing a vast and growing conduit for the distribution of artificially cheap and subsidized Chinese exports to the United States.

The U.S. relationship with China needs fundamental change: addressing the exchange rate policies and labor standards issues in the Chinese economy should be important national priorities. Wal-Mart’s huge reliance on Chinese imports illustrates that many powerful economic actors in the United States benefit from China’s unfair trading system. Wal-Mart’s gain, however, is not the country’s gain, as Wal-Mart’s imports have contributed to the ever-growing trade deficit that imperils future economic growth.

—The author thanks Josh Bivens and Ross Eisenbrey for comments; Elizabeth Glass for research assistance; and Molly McGrath, Kevin Rudiger, and Aditya Pande for data analysis.

About the author

Robert E. Scott is director of trade and manufacturing policy research at the Economic Policy Institute. He joined EPI as an international economist in 1996. Before that, he was an assistant professor with the College of Business and Management of the University of Maryland at College Park. His areas of research include international economics and trade agreements and their impacts on working people in the United States and other countries, the economic impacts of foreign investment, and the macroeconomic effects of trade and capital flows. He has a Ph.D. in economics from the University of California-Berkeley.

Endnotes

1. See Kimball and Scott (2014, 6 and “Appendix: Methodology,” 25–27) for further details.

2. This report distinguishes exports produced domestically and re-exports—which are goods produced in other countries, imported into the United States, and then re-exported to other countries, in this case to China. Re-exports do not support domestic employment because they are not produced domestically and they are excluded from the model used here. See Table 1 for information about the levels of U.S. re-exports to China in this period.

3. This model assumes that everything else is held constant; the trade and job loss estimates shown here are based on counterfactual simulations.

4. The complete list of Journal of Commerce citations for 2004–2015, covering calendar year trade between 2003 and 2014, is available on request.

5. Wal-Mart (2007) reports that it “estimates about $18 billion worth of products were purchased from China [in the fiscal year ending 2004] … about $9 billion imported from direct sources and about $9 billion from indirect.” These data are for Wal-Mart’s fiscal year ending on January 31, 2004, and were 11.9 percent of U.S. consumption imports from China in 2003, when most of those goods were imported. The following estimates all assume that Chinese imports are for Wal-Mart fiscal years (FY), and are compared with total U.S. imports in the preceding calendar years. Bianco and Zellner (2003) report that Wal-Mart imports from China totaled $12 billion (11.8 percent of U.S.-China imports) in FY 2002. Bianco (2006) reports that Wal-Mart imports from China were $22 billion in FY2005 (11.2 percent of China imports). Bianco’s estimates for FY 2004 replicate the estimate provided by Wal-Mart (2007) for its FY2004 imports from China. Based on these estimates, Table 1 assumes, conservatively, that Wal-Mart maintained a stable 11.2 percent share of U.S. goods imports from China between 2001 and 2013.

Between 2003 and 2013, overall Wal-Mart net sales in the United States rose from $208.8 billion to $336.6 billion (Wal-Mart Stores Inc. various years), rising from 7.7 percent of total U.S. personal consumption expenditures on goods in 2003 to 8.8 percent in 2013 (BEA 2015). Thus, Wal-Mart was a major and growing channel for the distribution of both domestic and imported goods in the United States in this period. Wal-Mart was also the single largest U.S. importer of goods imported from all countries via ocean container freight in 2014 (Journal of Commerce 2015), and was responsible for 12.1 percent of the total containers imported by the top 100 companies in that year. These data suggest that Wal-Mart’s share of total China imports likely increased between 2003 and 2013. Thus, the estimate of jobs displaced by Wal-Mart’s China trade in Table 1 likely represent a lower-bound estimate of actual jobs displaced.

6. Under U.S. rules, companies are allowed to petition Customs and Border Protection (CBP) to avoid disclosure of company names on bills of lading that accompany each shipment. Periodically, gaps appear in these disclosure petitions, making importing companies known for short periods of time. Comparable Wal-Mart data are available only for November 2007 and 2012 from this database.

7. This calculation is based on the ratio of total Wal-Mart international sales per square foot times an estimate of total Wal-Mart square footage in China, in various Wal-Mart fiscal years (Wal-Mart Stores Inc. 2002, 2006, 2014). Wal-Mart reports state that “over 95 percent of the merchandise in our stores in China is sourced locally” (Wal-Mart 2015b). Export estimates in this paper assume that sales per store in China were equal to the average per square foot for all Wal-Mart international stores times estimated total Wal-Mart square footage in China, and that all Wal-Mart imports into China came from the United States (the average Wal-Mart store in China was 2.3 to 2.8 times larger than the average Wal-Mart international store, based on data reported by Wal-Mart Stores Inc. (various years)). This is clearly an upper bound on total Wal-Mart exports to China because it assumes that all Wal-Mart imports into China originated in the United States, which is highly unlikely.

Wal-Mart had 6,107 international stores at the end of FY2014 and total international sales of $136.5 billion in 2014, or about $22.3 million per store. Wal-Mart had 405 stores in China, with estimated total sales of $25.6 billion in FY2014, and total imports of $1.0 billion (reported as U.S. exports in 2013 in Table 1). Assuming that all these imports were shipped from the United States, Wal-Mart was responsible for 0.9 percent of total U.S. exports to China in 2013.

8. These estimates assume that jobs supported and displaced by Wal-Mart’s China trade were directly proportional to total jobs supported and displaced by total U.S. exports to and imports from China in 2001 and 2013, as estimated by Kimball and Scott (2014).

9. Direct jobs displaced refer to jobs displaced within a given industry, such as motor vehicles and parts. Indirect jobs displaced are those displaced in industries that supply inputs to that industry, such as primary metal (e.g., steel), plastics and rubber products (e.g., tires and hoses), transportation, and information. Re-spending employment results from the spending of wages by employed workers. It is one form of a macroeconomic multiplier.

10. Author’s calculations from the estimated $1,800 wages lost by a median-wage non-college educated worker per year (Bivens 2013) times the 68.1 percent of the workforce made up of workers with less than a four-year college degree (U.S. Census Bureau 2015) times total number of U.S. workers employed (on average) in 2014 from the Bureau of Labor Statistics (BLS 2015) (yielding roughly 100 million non-college educated workers).

11. The initial Wal-Mart commitment was to purchase $50 billion in “U.S. products,” a figure that was subsequently increased to $250 billion (Loeb 2013, Walmart 2015a).

References

Alliance for American Manufacturing. 2015. “ANOTHER Walmart Made in America Infographic Needed Some Work, So We Fixed It.” Manufacture This (blog of the Alliance for American Manufacturing). July 7.

Bianco, Anthony. 2006. The Bully of Bentonville: How the High Cost of Wal-Mart’s Everyday Low Prices Is Hurting America. New York: Currency/Doubleday.

Bianco, Anthony, and Wendy Zellner. 2003. “Is Wal-Mart Too Powerful?” Business Week. October 3.

Bivens, Josh. 2013. Using Standard Models to Benchmark the Costs of Globalization for American Workers Without a College Degree. Economic Policy Institute, Briefing Paper #354.

Bureau of Economic Analysis (BEA). 2015. “Table 1.1.5 Gross Domestic Product.” (Accessed October 27).

Bureau of Labor Statistics (BLS). 2014. “Employment, Hours, and Earnings from the Current Employment Statistics Survey (National): Manufacturing Employment.” [Excel file].

Bureau of Labor Statistics (BLS). 2015. “Labor Force Statistics from the Current Population Statistics: Employment Level.” [Excel file].

Bureau of Labor Statistics, Employment Projections program (BLS–EP). 2014a. “Special Purpose Files—Employment Requirements Matrix; Chain-Weighted (2005 dollars) Real Domestic Employment Requirements Table for 2001” [Excel sheet, converted to Stata data file]. http://www.bls.gov/emp/ep_data_emp_requirements.htm

Bureau of Labor Statistics, Employment Projections program (BLS–EP). 2014b. “Special Purpose Files—Industry Output and Employment – Data for Researchers, Industry Output.” [CSV File, converted to Excel sheet and Stata data file].  http://www.bls.gov/emp/ep_data_industry_out_and_emp.htm

International Monetary Fund (IMF). 2015. International Financial Statistics. [CD-Rom, August 2015], Washington, D.C.: International Monetary Fund.

Journal of Commerce. 2015. “JOC Top 100 Importers in 2014: U.S. Foreign Trade Via Ocean Container Transport.” May 28.

Kimball, William, and Robert E. Scott. 2014. China Trade, Outsourcing and Jobs: Growing U.S. Trade Deficit with China Cost 3.2 Million Jobs between 2001 and 2013, with Job Losses in Every State. Briefing Paper #385. Washington, D.C.: Economic Policy Institute.

Loeb, Walter. 2013. “How Walmart Plans to Bring Manufacturing Back to the United States.” Forbes. November 12.

Panjiva.com. 2015. United States Trade Data (subscription data service). (Excel sheets accessed October 23).

Scott, Robert E. 2007. The Wal-Mart effect: Its Chinese imports have Displaced Nearly 200,000 U.S. Jobs. Issue Brief #235. Economic Policy Institute.

Scott, Robert E. 2013. Trading Away the Manufacturing Advantage: China Trade Drives Down U.S. Wages and Benefits and Eliminated Good Jobs for U.S. Workers. Briefing Paper #367. Economic Policy Institute.

Scott, Robert E. 2015. “Increased U.S. Trade Deficit in 2014 Warns Against Signing Trade Deal without Currency Manipulation Protection.” Economic Indicator: Trade and Globalization. Economic Policy Institute, February 5.

Sovereign Wealth Fund Institute (SWFI). 2015. “Sovereign Wealth Fund Rankings: Largest Sovereign Wealth Funds by Assets under Management.” Accessed October 27.

U.S. Census Bureau. 2013. “American Community Survey: Special Tabulation Over 45 industries, Covering 435 Congressional Districts and the District of Columbia (113th Congress Census Boundaries), Plus State and US Totals Based on ACS 2011 1-year file” [spreadsheets received March 6].

U.S. Census Bureau. 2015. “Current Population Survey, Historical Time Series, Table A-2: Percent of People 25 Years and Over Who Have Completed High School or College, by Race, Hispanic Origin and Sex: Selected Years 1940 to 2011” [Excel file].

U.S. International Trade Commission (USITC). 2014. “USITC Interactive Tariff and Trade DataWeb” [Excel files].

Wal-Mart Stores Inc. 2002. “Walmart 2002 Annual Report.”

Wal-Mart Stores Inc. 2006. “Walmart 2006 Annual Report.”

Wal-Mart Stores Inc. 2007. “Wal-mart Facts: Outsourcing.” [Html page downloaded April 3—available on request].

Wal-Mart Stores Inc. 2014. “Walmart 2014 Annual Report.”

Wal-Mart Stores Inc. 2015a. “Opportunity: US Manufacturing.”

Wal-Mart Stores Inc. 2015b. “Walmart China Factsheet.”

Wal-Mart Stores Inc. 2015c. “Walmart U.S. Manufacturing Announcements.”

Ongoing Joblessness in Texas: African American and Hispanic unemployment rates far exceed the white unemployment rate in the state

Ongoing Joblessness in New Mexico: Unemployment rate for Hispanics far exceeds the state’s white rate

Ongoing Joblessness in Michigan: Unemployment rate for African Americans tops in nation, more than double the state’s white rate

Ongoing Joblessness in Mississippi: Unemployment rate for African Americans ninth in nation, more than double the state’s white rate

Ongoing Joblessness in North Carolina: Unemployment rate for African Americans fourth in nation, more than double the state’s white rate

Yet, Social Security is the nation’s most successful anti-poverty program and it remains a fundamental pillar of the American economy—one that is critical to the long-term economic security of today’s young people. The new edition of A Young Person’s Guide to Social Security, released by the Economic Policy Institute and the National Academy of Social Insurance, gives young adults the information they need to participate in debates about Social Security’s future. The 60-page guide is written by young authors for students and young workers and explains why Social Security is not in grave danger as oft-reported.

The new edition coincides with the seminar for Washington interns, “Demystifying Social Security,” sponsored by NASI, and reflects the latest official estimates for Social Security in the 2012 Social Security Trustees’ report.

Note on citations: This textbook does not have a bibliography. All citations can be found in the PDF document as URLs inlaid in the text that link to the article or webpage from which the number, fact, or figure was derived.

All program statistics and historical data are from the Social Security Administration Office of the Chief Actuary and the Office of the Chief Actuary’s “2010 Trustees Report.” The most-used source is Appendix A of the report, Table VI.A4.— Operations of the Combined OASI and DI Trust Funds. All data, unless otherwise noted, refer to 2009.

Other sources include the Congressional Budget Office, the Employee Benefits Research Institute, and the Center for Retirement Research.

Event video (July 20, 2011):

Engaging younger generations in Social Security debates (Part 1)

(Part 2)

This briefing paper begins by providing background on the public sector’s commitment to equal opportunity and affirmative action in employment, and then explores the degree to which women and African Americans are overrepresented in state and local government jobs. It next turns to a discussion of how state and local public-sector workers have significantly higher levels of education than their private-sector peers, yet are consistently underpaid relative to similar private-sector workers. Then, it compares racial- and gender-based wage disparities in the state and local public sectors and the private sector. The briefing paper next explains the disproportionate impact of state and local public-sector job cuts on women and African Americans, and concludes by contrasting the private sector’s slow jobs recovery with continued employment declines in the public sector.

Key findings include:

• Historically, the state and local public sectors have provided more equitable opportunities for women and people of color. As a result, women and African Americans constitute a disproportionately large share of the state and local public-sector workforce.
• Overall, the wage gap across genders is similar in the state and local public sectors and in the private sector. However, it is smaller for highly educated women employed in state and local government.
• State and local public-sector workers of color face smaller wage disparities across racial lines, and at some levels of education actually enjoy a wage premium over similarly educated white workers.
• The disproportionate share of women and African Americans working in state and local government has translated into higher rates of job loss for both groups in these sectors. Between 2007 (before the recession) and 2011, state and local governments shed about 765,000 jobs. Women and African Americans comprised about 70 percent and 20 percent, respectively, of those losses. Conversely, Hispanic employment in state and local public-sector jobs increased during this period (although most of that increase likely occurred in the lowest-paid jobs).
• Job losses in the state and local public sectors stand in contrast to the jobs recovery in the private sector. From February 2010 (the month the labor market “bottomed out”) to January 2012, the United States experienced a net increase in total nonfarm employment of more than 3.2 million jobs, while state and local government employment fell by 438,000. Over this period, every major sector of the economy experienced net growth in jobs except the public sector.

The public sector’s commitment to equal opportunity and affirmative action

In the 1960s and 1970s, the federal government, through a combination of executive orders and legislation, prohibited discrimination on the basis of sex and race in employment and the payment of wages.1 Studies of the hiring practices and wages of the state and local public sectors have shown the effectiveness of anti-discrimination policies, especially in contrast to the private sector. Since the creation of equal opportunity and affirmative action programs, women and African Americans have seen greater employment opportunities in the economy as a whole, but particularly in the public sector (Crosby 2004). Though discrimination in the public sector likely still exists,2 government remains a model of how to achieve greater equality in employment and workplace diversity.

While some would argue that the United States’ labor market today is largely free of prejudice and discrimination, a substantial and growing body of research suggests that gender- and race-based prejudices continue to afflict the U.S. workforce.3 These prejudices often take the form of wage disparities. Today, women earn only 77 cents for every dollar paid to their male counterparts, and the situation is worse for African American and Hispanic women, who earn only 62 cents and 54 cents, respectively, for every dollar paid to their non-Hispanic white male counterparts (National Women’s Law Center 2012).4 Furthermore, the Equal Employment Opportunity Commission continues to win settlements against employers in race discrimination cases based on compensation disparities.5

Research buttresses this evidence of wage discrimination with findings of significant race- and gender-based discrimination in hiring. For example, Harvard University researchers found that résumés with “white-sounding” names such as “Emily” are 50 percent more likely to elicit interviews than equivalent résumés with “black-sounding” names such as “Lakisha” (Bertrand and Mullainathan 2004). In addition, a multi-year, national study on race and sex discrimination in large and midsized private businesses found that intentional discrimination exists in every region of the country and in each of nine occupational categories, and it “is so pervasive that affirmative action programs continue to be necessary” (Blumrosen and Blumrosen 2002). Even as recently as this year, the U.S. Department of Labor Office of Federal Contract Compliance Programs found that FedEx engaged in discrimination against 21,000 applicants in 15 states (U.S. Department of Labor 2012). In short, although the American ideal may be to judge individuals by the content of their character, we have not yet guaranteed equal opportunity in all cases.

Today, every job in state and local government is subject to federal regulations concerning equal opportunity (Dale 2005), and many state and local governments require affirmative action plans beyond federal equal opportunity requirements. When compared with the private sector, the state and local public sectors have gone to greater lengths to enact affirmative action policies. However, many of the affirmative action programs implemented by state and local governments have met opposition from state legislatures and governors proposing to ban such laws.6

In the private sector, affirmative action laws and regulations are comparatively few. Federal law requires only two types of private-sector employers to implement affirmative action plans: those that have federal contracts or subcontracts in excess of $50,000 and that also have at least 50 employees, and those with 15 or more employees that have faced a judicial finding of discrimination.

Tallying the number of public- and private-sector jobs subject to monitoring requirements and set-aside programs, about one in four American workers hold jobs covered by mandatory federal affirmative action programs (U.S. Department of Labor 2002).

Despite the persistence of discrimination in state and local government, affirmative action and equal opportunity policies have transformed the public sector, relative to the private sector, into increasingly hospitable employers of women and African Americans. Fifty years of efforts to redress past discrimination have proven their effectiveness in greater numbers of women and African Americans entering state and local government. As a result, public-sector jobs at the state and local levels remain critical to their livelihoods.

The importance of the state and local public sectors to women and African Americans

Figure A

For decades, women and African Americans have been employed in the state and local public sectors at rates that are higher than their shares of private-sector and overall employment. As shown in Figure A, in 2011 women comprised 48.3 percent of overall employment, yet accounted for 59.5 percent of employment in state and local government, significantly higher than their 46.7 percent share of private-sector employment. As the figure illustrates, women’s share of state and local government jobs has increased by 3.3 percentage points since 1989.

Figure B

In comparison, Figure B illustrates that in 2011, African Americans accounted for 10.9 percent of overall employment, yet held 12.8 percent of state and local public-sector jobs and 10.3 percent of private-sector jobs. As the figure shows, African Americans, like women, have traditionally been underrepresented in the private sector and overrepresented in the state and local public sectors. However, African Americans’ representation in state and local government jobs has declined 2.3 percentage points since 1997.

Figure C

In contrast to the patterns among women and African Americans, Hispanics remain underrepresented in the state and local public sectors and overrepresented in the private sector. As illustrated in Figure C, in 2011 Hispanics made up 15 percent of overall employment, yet accounted for 10.6 percent of state and local government employment, far lower than their 15.8 percent share of private-sector employment. The figure shows how the Hispanic share of employment in state and local government and in the private sector has steadily kept pace with Hispanics’ growth in overall employment since 1989.

A better-educated workforce, lower overall pay

Figure D

When making any wage comparisons across the public and private sectors, it is important to recognize that workers in the state and local public sectors have, on average, a different demographic profile than workers in the private sector. As previously noted, state and local government workers are more often female, and a greater share is African American. Figure D further illustrates these patterns by graphing the share of employment for state and local public-sector employees and private-sector employees by gender and race.

Figure E

State and local public-sector employees also tend to have significantly higher levels of education than private-sector workers. Figure E compares the education levels of state and local public-sector workers with those of private-sector workers, separated by sex. As the figure shows, 46.2 percent of men in state and local government jobs have at least a bachelor’s degree, compared with 29.1 percent in the private sector. For women, the difference is even larger: 54.1 percent in the state and local public sectors have at least a bachelor’s degree, compared with 30.1 percent in the private sector.

Figure F

Figure F makes a similar comparison of education levels in the state and local public sectors versus the private sector, this time separated by race. State and local government workers show significantly higher levels of education across all racial groups, with particularly striking differences for African Americans and Hispanics. For African Americans, the share with at least a bachelor’s degree, at 42.1 percent, is more than double that of the private sector (20.1 percent). Among Hispanic state and local government workers, 34 percent have at least a bachelor’s degree—nearly three times the 11.7 percent share in the private sector.

Despite these significantly higher levels of education—and contrary to assertions by some governors in recent state-level debates—the most rigorous studies have consistently shown that state and local government employees earn less both in wages and total compensation than comparable private-sector workers (Keefe 2010). Using data from the Annual Social and Economic Supplement of the Current Population Survey and standard regression models for wage analyses, we compared the wage income of private-sector employees with that of state and local government workers. After controlling for education, experience, sex, race, ethnicity, marital status, full-time/part-time status, number of hours worked, citizenship status, Census region, metropolitan status (whether residing within or outside the boundaries of a major metropolitan area), and employer size, we find that state and local government employees make, on average, 11.7 percent less in wages than similar private-sector employees.7 (These same controls are used in all subsequent wage comparisons in this briefing paper.) Other studies looking at total compensation including employer-provided benefits find a narrower gap but that public-sector workers are still under-compensated in comparison to private-sector workers (Keefe 2010, 2011).

Table 1

As also shown in Table 1, a gap between state and local public-sector workers and similar private-sector employees appears for both genders and for whites, African Americans, and Hispanics. In a regression that allowed for the effect of state and local government employment on wages to differ between men and women,8 there was no statistically significant difference between the sexes: Both earned about 11–12 percent less than private-sector workers of the same gender.

An examination of how the state and local public-sector wage penalty applies to individuals of different races reveals that whites in state and local government jobs earn 14.5 percent less than whites in the private sector. The results indicate smaller wage gaps for African American and Hispanic workers, with these groups earning 1.7 percent and 3.7 percent less, respectively, than private-sector employees of the same race. It should be noted that these smaller wage gaps do not stem from people of color in state and local government jobs earning more than their white colleagues; overall, they earn less. Rather, the large racial wage gap for African Americans and Hispanics that exists in the private sector offsets much of the state and local public-sector wage penalty (a finding that will be discussed in greater depth in the following section).

A smaller wage disparity for many state and local government workers

Having established that a wage penalty exists for state and local public-sector employees when compared with private-sector workers regardless of race or gender, we now examine how race- and gender-based wage disparities in state and local government compare with those in the private sector. For women with at least a bachelor’s degree—who, as mentioned earlier, account for 54.1 percent of women in the state and local public sectors—the wage gap with similarly educated males is smaller in state and local government than in the private sector. Likewise, the wage disparities between whites and African Americans, and between whites and Hispanics, are significantly smaller in the state and local public sectors. This, combined with the model hiring practices previously described, may help explain why the public sector continues to attract disproportionate shares of women and African Americans.

Table 2 shows the male-female wage gap in state and local government and in the private sector. Among state and local public-sector workers, women on average earn 20.9 percent less in wages than their male counterparts. This is statistically no different from the average male-female wage gap of 20.4 percent in the private sector.9 However, as previously noted, state and local government employees tend to have significantly higher levels of education than private-sector workers, particularly among women. The data show that the state and local public sectors do a better job of equalizing pay across genders for workers with higher levels of education. Women with a bachelor’s degree in state and local government jobs earn 16.9 percent less in wages than their male counterparts, compared with a male-female wage gap of 18.9 percent for similarly educated private-sector workers. The benefit of working in state and local government jobs is more striking for women with advanced degrees. While they still earn 12.4 percent less than similarly educated men in state and local public-sector jobs, this is far smaller than the private-sector wage gap of 21 percent.

Table 2

State and local government jobs also strongly reduce the wage gap between different racial groups. Table 3 shows differences in wages for African Americans and Hispanics compared with white workers in the private sector and in the state and local public sectors. As the table shows, the difference between the sectors is dramatic. In the private sector, African Americans earn an average of 12.9 percent less than white workers. Yet among state and local public employees, the wage disparity between African Americans and whites is only 2.2 percent. Likewise, Hispanic workers in the private sector earn, on average, 11.1 percent less than white workers. In the state and local public sectors, this disparity is only 2.9 percent.

Table 3

For African Americans and Hispanics, public-sector jobs demonstrate a lower racial wage gap at some education levels, while demonstrating a wage premium at others. Table 4 shows how the wages of African Americans and Hispanics with different levels of education compare with wages of white workers with equivalent education levels. In the private sector, African Americans at every education level earn wages significantly lower than those of white workers. The smallest wage gap in the private sector is for African Americans with some college, who earn 11 percent less than similarly educated whites. The largest difference is for African Americans with advanced degrees, who earn an average of nearly 20 percent less than whites with advanced degrees. In contrast, in state and local government jobs, African Americans at some education levels receive higher wages than similarly educated whites. For example, African Americans with less than a high school degree, a bachelor’s degree, or an advanced degree earn 8.3 percent, 0.4 percent, and 1.7 percent more, respectively, than whites with the same levels of education. Still, the majority of African Americans in state and local public-sector jobs have either a high school degree or some college. For these two groups, wages are 5.7 percent and 5.2 percent lower, respectively, than those of similarly educated whites.

Table 4

Hispanic workers also benefit from more equitable wages in the state and local public sectors. The private-sector wage gap between Hispanics and whites ranges from a low of 7.4 percent for workers with less than a high school degree to a high of 19.6 percent for workers with an advanced degree. In contrast, in state and local government jobs, Hispanic employees with less than a high school education earn 6.7 percent more than similar white employees, and Hispanic workers with advanced degrees earn only 2 percent less than similarly educated whites.

Public-sector job loss: Disproportionately harmful to African Americans and women

The high concentration of women and African Americans working in the public sector, and the greater wage equity many experience there, make cuts to state and local government especially painful for both groups. The Great Recession caused the largest drop in state revenues ever recorded and left many states facing dramatic budget shortfalls (McNichol et al. 2012). Because most state constitutions do not allow deficit spending, this has led to steep reductions in state and local budgets, which has translated into significant job loss among state and local public-sector employees.

Reductions in state and local government workforces are a significant drag on the economy as a whole (Leonhardt 2011)—and are particularly damaging for women. As previously discussed, in 2011 women comprised 48.3 percent of overall employment, but about three-fifths (59.5 percent) of state and local public-sector workers. The disproportionate representation of women in state and local government has resulted in women suffering the vast majority of public-sector job losses. Of the net change in total state and local government employment between 2007 (before the recession) and 2011—a decline of roughly 765,000 jobs—about 70.5 percent of the jobs lost were held by women (Current Population Survey Outgoing Rotation Group 2007–2011). Today approximately 540,000 fewer women are employed in state and local government jobs than in 2007, compared with about 225,000 fewer men (Current Population Survey Outgoing Rotation Group 2007–2011). These numbers represent declines of 5.1 percent and 3.2 percent, respectively (see Figure G).

Figure G

African Americans have also suffered disproportionately from state and local budget cuts. As noted previously, in 2011 African Americans comprised more than one-tenth (10.9 percent) of overall employment and 12.8 percent of state and local public-sector employment. However, they accounted for almost one-fifth (19.8 percent) of the overall decline in state and local government employment between 2007 and 2011 among racial groups that lost jobs (Current Population Survey Outgoing Rotation Group 2007–2011). This loss of 177,000 jobs (Current Population Survey Outgoing Rotation Group 2007–2011) represents a decrease in African Americans’ state and local government employment of 7.6 percent. As shown in Figure G, this is the largest percentage change for all racial groups.

Examining the proportion of people from the state and local public sectors who are currently unemployed provides another perspective on public-sector job cuts. In 2011 nearly 450,000 women reported that they were unemployed and that their most recent job was from the state or local public sectors (see Figure H). At 62 percent of the total number of people unemployed from the state or local public sectors, this is lower than women’s share of the net change in state and local government jobs (as noted previously, about 70.5 percent)—suggesting that some of the women who lost state and local public-sector jobs since the recession began have either found private-sector work or have exited the labor force (i.e., retired or stopped looking for a job). Nevertheless, it is still larger than the overall female share of state and local government employees.

Figure H

The situation for African Americans is the reverse. In 2011, African Americans comprised 26.9 percent of those reporting that they were unemployed and had most recently worked in state or local government jobs (Current Population Survey Outgoing Rotation Group 2011). This is much higher than their share (19.8 percent) of the overall decline in state and local public-sector employment among racial groups that lost jobs from 2007 to 2011. This finding suggests that, unlike other groups who either took jobs in the private sector or exited the labor force since the beginning of the recession, African Americans have faced greater difficulty in finding other work and/or remained more strongly attached to the labor market, leading to their higher share of those still unemployed.

Although women and African Americans have experienced significant declines in state and local public-sector employment, Hispanic employment in these sectors has actually increased since 2007. Total Hispanic employment in state and local government jobs grew by about 107,000 people from 2007 to 2011, an increase of 6.3 percent (Current Population Survey Outgoing Rotation Group 2007–2011). As noted previously, in 2011 Hispanics made up 10.6 percent of all state and local public employees—up from 9.6 percent in 2007. This increase in Hispanic workers in state and local government jobs is a positive step toward achieving greater racial and ethnic diversity in employment; however, it accompanies less positive wage trends. Since the start of the recession, the real median wage of Hispanic employees in state and local public-sector jobs has declined by 5.2 percent, compared with a decline of 1.9 percent for African Americans, 0.7 percent for whites, and 2.2 percent for all other races (Current Population Survey Outgoing Rotation Group 2007–2011). This suggests that although more Hispanic workers are entering into state and local public-sector jobs, they are largely taking lower-paying positions.

The private sector’s slow improvement and the public sector’s continued decline

Table 5

Since the recession’s official end in June 2009, job growth in the United States has been distinctly one-sided: The private sector has experienced employment growth in the majority of industries, while the public sector has continued to shed jobs. Table 5 details changes in employment by sector from June 2009 to January 2012. Over this period, the United States saw a net increase in nonfarm employment of roughly two million jobs, with every non-government sector adding jobs except construction and financial services. Over the same period, however, state and local government employment shrunk by nearly 580,000 jobs. It is even more telling to look at the period after February 2010, the month the labor market “bottomed out.” From February 2010 to January 2012, the United States experienced a net increase in total nonfarm employment of more than 3.2 million jobs, while state and local government employment fell by 438,000. Over this period, every major sector of the economy experienced net growth in jobs except the public sector.

Conclusion

For the past five decades, the public sector has led the way in providing opportunity and reducing discrimination in the workforce. This has led to the disproportionate representation of women and African Americans in state and local government jobs. Unfortunately, this overrepresentation has meant that, as state and local governments address revenue shortfalls by slashing budgets and cutting public services, women and African Americans have suffered disproportionately from the resulting job losses since 2007.

The continued cuts to state and local governments also threaten to undermine progress that the public sector has made toward greater wage equality. The economy is losing jobs in a sector (state and local government) that often has smaller pay gaps than the private sector. Especially for people of color and women with high levels of education, this is a step in the wrong direction.

With the private sector finally showing signs of sustained job growth, continued cuts to state and local governments only hamper a faster recovery. Absent further federal assistance and an altered approach to raising state and local revenues, the state and local public sectors will likely continue to shed jobs. An expanded federal recovery program—such as greater fiscal relief to states, funding for infrastructure and school modernization projects, continued support of social insurance programs, and direct job creation programs in hard-hit communities—would go a long way toward accelerating the recovery and assisting women and African Americans who have suffered disproportionately from state and local public-sector job cuts.

Endnotes

1. The establishment and strengthening of civil rights laws protecting women and people of color in employment occurred through the Equal Pay Act of 1963, Federal Executive Order 11246, and Titles VI and VII (as amended by the Equal Employment Opportunity Act of 1972) of the Civil Rights Act of 1964 (U.S. Equal Employment Opportunity Commission).

2.  Police departments have historically lagged far behind other government agencies in hiring women (Crosby 2004), yet have hired African American men in proportion to their labor force participation rate (Austin 2011). In contrast, fire departments have imposed barriers to hiring and promoting African American men, who remain underrepresented since they are employed at only 67 percent of their proportionate representation level (Austin 2011).

3.  It is beyond the scope of this paper to provide a full review of this literature. A good starting point would be Riach and Rich 2002. In addition to field experiments, there are also implicit bias studies (see Grant-Thomas 2011 for an introduction), interviews of employers (Moss and Tilly 2001), and the fact that multivariate regression-based analyses fail to fully explain wage differentials (General Accounting Office 2003).

4.  National Women’s Law Center calculations from U.S. Census Bureau, Current Population Survey, 2011 Annual Social and Economic Supplement, Table PINC-05: Work Experience in 2010 – People 15 Years Old and Over by Total Money Earnings in 2010, Age, Race, Hispanic Origin, and Sex, available at http://www.census.gov/hhes/www/cpstables/032011/perinc/toc.htm (last visited Sept. 27, 2011). Annual racial wage gaps were calculated by subtracting the annual total earnings of African American and Hispanic women from that of non-Hispanic white men.

5.  In EEOC, et al. v. KOKH, No. 5:07-cv-01043-D (W.D. Okla. March 4, 2011), a television station settled a race and sex discrimination case filed by the EEOC for $45,000 and additional consideration. In EEOC v. Williams Country Sausage Co., Civil Action No. 1:10-cv-01263 (W.D. Tenn. filed Sept. 30, 2010), a pork company settled a race discrimination case filed by the EEOC for $60,000 and other relief.

6.  Opponents of state affirmative action efforts have sought to ban those policies through state ballot initiative. For example, in 1996, voters approved California’s Proposition 209, which amended the state constitution to overturn affirmative action policy in state public employment, public education, and public contracting. Court cases on affirmative action, including Fisher v. University of Texas (to be heard by the U.S. Supreme Court this year), have focused largely on state public university admissions policies.

7.  EPI analysis of Annual Social and Economic Supplement of the Current Population Survey, pooled years 2006 and 2007. We use these years because they are the most recent normal business cycle, thereby eliminating any biases in the data resulting from the effects of the subsequent recession.

8.  Allowing for the effect of state and local government employment on wages to vary by gender provides for the possibility that only men or only women receive lower wages in the state and local public sectors when compared with their private-sector counterparts. The results demonstrate that this is not the case.

9.  In the regression model in which a state and local public-sector indicator variable interacts with the female indicator variable (model two in Table A1), the female state and local interaction term is not statistically different from zero. This indicates that the male-female wage gap is, on average, no different across the two sectors.

Table and figure notes

References

Austin, Algernon. 2011. Whiter Jobs, Higher Wages: Occupational Segregation and the Lower Wages of Black Men. Economic Policy Institute Briefing Paper No. 288.

Bertrand, Marianne, and Senhil Mullainathan. 2004. “Are Emily and Greg More Employable Than Lakisha and Jamal? A Field Experiment on Labor Market Discrimination.” The American Economic Review, vol. 94, no. 4, pp. 991–1013.

Blumrosen, Alfred W., and Ruth G. Blumrosen. 2002. The Reality of Intentional Job Discrimination in Metropolitan America—1999. Rutgers State University of New Jersey.

Crosby, Faye J. 2004. Affirmative Action Is Dead; Long Live Affirmative Action. Yale University Press.

Current Establishment Survey. 2009–2012. Survey conducted by the Bureau of Labor Statistics. U.S. Census Bureau.

Current Population Survey Annual Social and Economic Supplement microdata. Various years. Survey conducted by the Bureau of the Census for the Bureau of Labor Statistics [machine-readable microdata file]. U.S. Census Bureau. http://www.bls.census.gov/cps_ftp.html#cpsmarch

Current Population Survey Outgoing Rotation Group microdata. Various years. Survey conducted by the Bureau of the Census for the Bureau of Labor Statistics [machine-readable microdata file]. U.S. Census Bureau. http://www.bls.census.gov/cps_ftp.html#cpsbasic

Dale, Charles V. 2005. Congressional Research Service Report for Congress: Federal Affirmative Action Law: A Brief History. Congressional Research Service, Library of Congress.

Equal Employment Opportunity Act of 1972, Pub. L. 92-261. 86 Stat. 103. 24 Mar. 1972. Print.

General Accounting Office. 2003. Women’s Earnings: Work Patterns Partially Explain Difference between Women’s and Men’s Earnings. General Accounting Office.

Grant-Thomas, Andrew. September/October 2011. “Implicit Bias, Racial Inequality, and Our Multivariate World.” Poverty & Race. Poverty & Race Research Action Council.

Keefe, Jeffrey. 2010. Debunking the Myth of the Overcompensated Public Employee: The Evidence. Economic Policy Institute Briefing Paper No. 276.

Keefe, Jeffrey. 2011. Research series on public- versus private-sector compensation in the states, published by the Economic Policy Institute and accessible through through EPI’s “Public-sector workers earn less” Web commentary. http://www.epi.org/publication/public_sector_workers_earn_less/

Leonhardt, David. 2011. “The Cost of Austerity.” The New York Times’ Economix blog, July 8. http://economix.blogs.nytimes.com/2011/07/08/the-cost-of-austerity/

McNichol, Elizabeth, Philip Oliff, and Nicholas Johnson. 2012. “States Continue to Feel Recession’s Impact.” Center on Budget and Policy Priorities, updated February 27, 2012. http://www.cbpp.org/cms/index.cfm?fa=view&id=711

Moss, Philip, and Chris Tilly. 2001. Stories Employers Tell: Race, Skill, and Hiring in America. Russell Sage Foundation.

National Women’s Law Center. 2012. Closing the Wage Gap is Especially Important for Women of Color in Difficult Times. http://www.nwlc.org/sites/default/files/pdfs/womenofcolorfactsheet.pdf.

Riach, P.A., and J. Rich. 2002. “Field Experiments of Discrimination in the Marketplace.” The Economic Journal, vol. 112, no. 483, pp. F480–F518.

U.S. Department of Labor, Office of Federal Contract Compliance Programs. 2002. “Facts on Executive Order 11246—Affirmative Action.” Accessed March 30, 2012, at http://www.dol.gov/ofccp/regs/compliance/aa.htm.

U.S. Department of Labor, Office of Federal Contract Compliance Programs. 2012. “Shipping Giant FedEx to Pay $3 Million to Settle Charges of Hiring Discrimination Brought by U.S. Department of Labor.” Accessed March 30, 2012, at http://www.dol.gov/opa/media/press/ofccp/OFCCP20120507.htm.

U.S. Equal Employment Opportunity Commission. n.d. “Laws Enforced by EEOC.” Accessed March 2, 2012, at http://www.eeoc.gov/laws/statutes/index.cfm.

Appendix tables

Table A1

Table A2

This finding, and previous research by the same authors (Biggs and Richwine 2011), are at odds with a large body of research showing that public school teachers and other government workers have total compensation that is lower—or at least no higher—than that of comparable private-sector workers (see, for example, Allegretto, Corcoran, and Mishel 2004, 2008, 2011; Bender and Heywood 2010; Keefe 2010; Munnell et al. 2011; Schmitt 2010). Furthermore, the “teaching penalty” has grown, as teachers’ and other public-sector workers’ pay has declined relative to that of comparable private-sector workers (Allegretto, Corcoran, and Mishel 2008, 2011; Bender and Heywood 2010).

How do Richwine and Biggs get such different results? Their research comparing public- and private-sector pay has been critiqued elsewhere (Hanauer 2011; Keefe 2011), so this brief will address the authors’ specific case against school teachers, focusing on benefits. A separate EPI paper on teacher salaries by Mishel and Roy (forthcoming), as well as an overview by Jeffrey Keefe in the National Education and Policy Center’s Think Tank Review Project (2012), challenge Richwine and Biggs’ extremely controversial claim that teacher salaries are as high as those of comparable workers, once teachers’ supposed “low cognitive ability” compared to other college graduates is taken into account.

This brief will show that, among other things, Richwine and Biggs:

• compare teachers with private-sector workers with much lower educational attainment
• selectively alternate between the cost of benefits to employers and the value to workers, and inappropriately equate the latter with the often much higher cost to individuals of obtaining equivalent benefits
• triple the cost of teacher pensions by assuming a very low rate of return on pension fund assets, or by assuming a very high cost of guaranteeing these benefits
• inflate the cost of retiree health benefits and seasonal leave
• place an arbitrary dollar figure on the value of job security to workers, while ignoring the advantage to employers of employee retention.

What is the appropriate comparison group?

There is near unanimity among economists that, whenever possible, compensation comparisons should compare workers with the same years of education and experience, with age often used as a proxy for experience. It is thus worth pointing out that Richwine and Biggs—despite their protestations to the contrary—generally compare teachers and other public-sector workers with private-sector workers who might be expected to have lower salaries or less-generous benefits. As Mishel and Roy (forthcoming) explain, Richwine and Biggs highlight statistical results for wages that intentionally exclude education. Though they include education and a range of other demographic variables in their initial wage regression, they later inexplicably drop the education variable. Likewise, in comparing benefits, they compare public school teachers with all private-sector workers employed by large employers, even though public school teachers are much better educated than these private-sector workers.

Gender, race, marital status, and employer size are also significant predictors of compensation—though there is no clear-cut economic rationale for why women, minorities, and unmarried people should be paid less, nor why larger employers should pay more. Nevertheless, it is conventional to include these demographic variables as well as employer size where possible.

Richwine and Biggs imply that they are bending over backward to compare public-sector workers to better-paid private-sector workers. However, researchers, including Richwine and Biggs, are if anything minimizing the overall pay gap by including race and gender controls in wage regressions, which may explain away some of the lower pay of teachers and other public-sector workers, who are more likely to be female and black. Schmitt (2010), for example, finds that while state and local workers are paid 6.4 percent less than private-sector workers with the same education and age, this pay gap shrinks to 3.7 percent if controls for gender, race, and region are included.

Philosophical issues aside, the ability to make a clear, apples-to-apples comparison of the generosity of benefits is limited by the data. Researchers often compare wages using the Census Bureau’s Current Population Survey (CPS), which includes detailed demographic information as well as firm size (though Richwine and Biggs actually omit firm size in their wage regression). Research on fringe benefits, on the other hand, is usually based on the Bureau of Labor Statistics’ National Compensation Survey (NCS), an employer survey that only provides breakdowns by broad occupation group, industry group, and establishment size.

Since there is no way to directly compare the benefits of workers with the same education and experience, researchers sometimes compare teachers and other public-sector workers to private-sector workers in large establishments, based on the fact that most public-sector workers also work for large employers, which enjoy cost advantages in providing fringe benefits. This is the approach adopted by Richwine and Biggs. Though this may be the least-bad option, Richwine and Biggs falsely claim that they are being “relatively conservative” in choosing this comparison group, even though public school teachers are better educated and therefore better paid than private-sector workers working for large employers—and correspondingly might be expected to have more-generous benefits (which is not surprising when you consider that the comparison group includes Walmart workers, for example).

Public school teachers are very well educated, split roughly evenly between those with bachelor’s degrees and those with graduate degrees (49 percent have a bachelor’s degree, and 45 percent have at least a master’s degree). They are much better educated than private-sector workers with large employers. For example, among private-sector workers working for firms with 1,000 or more employees, only 21 percent have a bachelor’s degree, and just 9 percent have at least a master’s degree (EPI analysis of 2003–2010 IPUMS CPS data [King et al. 2010]).1

How Richwine and Biggs get to 52 percent

Despite these differences in educational attainment, Richwine and Biggs acknowledge at the outset that National Compensation Survey data show that the relative importance of benefits for teachers and private-sector workers in large establishments is nearly identical, amounting to roughly 41 percent of wages (or 29 percent of total compensation). However, after making various adjustments, Richwine and Biggs estimate that the value of teachers’ benefits is actually more than double the NCS estimate, or equal to 100.8 percent of wages (they also make a much smaller revision to the estimate for private-sector workers). As shown in Table 1, this is achieved by almost tripling retirement costs; by adding a benefit missing from the Bureau of Labor Statistics data (retiree health care); and by adding a benefit they call “work-year leave” that is already factored into wage measures.

Table 1

In addition, Richwine and Biggs inflate teachers’ total compensation by an additional 8.6 percent to account for teachers’ supposedly greater job security. Since, as mentioned earlier, they also adjust teachers’ salaries upward to match those of private-sector workers, teachers’ total compensation is supposedly 52 percent higher than that of private-sector workers.2

However, as will be discussed in this paper, the only one of these adjustments that is relatively uncontroversial is the addition of retiree health benefits (though even this figure appears inflated). Furthermore, the addition of retiree health care does not by itself close the pay gap between public school teachers and comparable private-sector workers, which the authors initially estimate at 19.3 percent (as mentioned earlier, their later assumption that there is no salary gap rests on a highly dubious model that excludes education).

Valuing benefits: The cost to employers vs. the value to workers

Richwine and Biggs selectively alternate between the cost of benefits to employers and the value to workers. They inappropriately equate the latter with the often much higher cost to individuals of obtaining equivalent benefits.

When assessing the value of fringe benefits, researchers may be interested in the cost to employers (with or without taking into account indirect costs and benefits, such as employee retention); the value to employees; or the cost of purchasing similar benefits in the private market. Large private employers and the government are often able to provide insurance and pension benefits at much lower cost than individuals can purchase the same benefits for themselves, which is one of the main reasons these benefits are provided by employers in the first place (there may also be tax advantages as well as workforce management considerations).

Large employers and government entities are better equipped to assume many risks than are individuals because individual longevity and medical risks can be reduced or even eliminated by pooling. This is a basic tenet of insurance: Costs fall as pool size increases. This asymmetry, plus economies of scale in administration and the elimination of adverse-selection problems that drive up costs in the individual insurance market (because insurers assume that sicker individuals are more likely to purchase health and life insurance and healthier individuals are more likely to purchase life annuities), make employer-provided pension and insurance benefits a very cost-effective component of employee compensation for large employers. That is, the value to workers is greater than the cost to employers.

Though the value to workers is generally higher than the cost to employers, this does not mean that the value to employees is the relevant measure, nor that the value to employees is the same as the cost of purchasing benefits in the individual insurance market, as Richwine and Biggs seem to suggest. Many workers would purchase less-generous benefits or forgo them entirely if required to purchase them at the higher cost.

It is far from clear that it is the value of benefits to workers that is of interest in this case. Since the main point of Richwine and Biggs’ paper is that taxpayers are being “overcharged,” the direct cost to employers would seem to be the most straightforward measure. This is what most researchers focus on, including Richwine and Biggs—though not consistently. For example, they arbitrarily value retiree health benefits (which are more common in the public sector) at the higher cost of purchasing equivalent insurance in the private market. They do not, however, make the same adjustment for group health insurance for active workers, nor for disability and life insurance provided through Social Security. (Since many teachers are not covered by Social Security, the latter adjustment would tend to increase private-sector pay relative to teacher pay.) Richwine and Biggs do not explain this inconsistency, giving the impression that they choose among measures with an eye to inflating teacher compensation.

The perceived value of benefits to workers may be relevant if, say, certain benefits are valued more than others relative to their direct cost, and if this affects recruitment and retention. However, this raises the thorny issue of how to measure the value to workers, since by definition workers are not directly paying for employer-provided benefits (workers’ own contributions toward retirement and other benefits are not included in these compensation measures). In any case, the value to workers is not the same as the cost of purchasing equivalent benefits in the individual market, which is irrelevant.

This also raises the issue of why public-sector employers would take greater advantage of the difference between the direct cost of providing certain benefits and their (presumably higher) value to workers. Advocates of high-road employment practices and social insurance might argue that the full value of employer- and government-provided benefits is not fully recognized by private-sector employers. However, this is an odd position coming from the Heritage Foundation and the American Enterprise Institute, organizations that tend to view the private sector as more efficient, and favor shifting the cost and risk of retirement and health care to individuals.

The present value of future benefits

The problem of distinguishing between the cost to employers and the value to workers is compounded in the case of future retiree benefits by the fact that these benefits are uncertain and that they must be translated to present values. By selectively focusing on the supposed value to workers rather than the direct cost to employers, and by placing a very high value on pension guarantees, Richwine and Biggs value pension benefits at triple their direct cost to employers.

Admittedly, valuing retiree benefits is a complicated task. First, future benefits must be estimated, and these estimates are sensitive to underlying assumptions. The future value of pension benefits, for example, depends in part on salary projections, because service credits are usually multiplied by a percentage of final pay rather than current pay. Similarly, the future value of retiree health benefits depends not only on projected health care costs, but also on whether the benefits will even exist when workers retire. (Unlike accrued pension benefits, which are generally protected by law, retiree health benefits may be reduced or eliminated at any time, with the possible exception of those covered by a collective bargaining agreement.)

Second, future benefits must be translated into present values. In the case of retiree benefits that are funded in advance (including traditional pension benefits and some retiree health benefits), this is equivalent to asking how much employers need to contribute to a trust today to pay for benefits in the future, which depends on investment returns.

Pension benefits

The problem of translating future benefits to present values looms especially large in the case of pension benefits. Importantly, Richwine and Biggs do not directly challenge pension fund actuaries’ assumptions about expected returns, which are generally slightly lower than the returns these funds have realized historically. Rather, they argue that since returns are uncertain, the yardstick should be the so-called risk-free rate—the long-run return on Treasury securities—which is roughly half the expected return on pension fund assets.  Due to compounding, investing in low-yield assets such as Treasury bonds would triple required pension contributions by Richwine and Biggs’ estimation.

Most economists, including Richwine and Biggs, agree that the risk-free rate is much lower than the expected return on actual pension fund assets, which are invested in balanced portfolios that include stocks. Neither Richwine nor Biggs denies the existence of an equity premium (a higher expected return on stocks than the risk-free rate), and Biggs has been bullish on stock returns in other contexts (Biggs 2002). However, Richwine and Biggs point out that economists would use the risk-free rate, rather than the expected return, to determine how much employers or workers would need to set aside to guarantee a similar retirement benefit.

The question boils down to which measure is appropriate in this context. As with benefits for active workers, the issue can be framed as the difference among the direct cost to employers, the total (direct and indirect) cost to employers net of any benefits received, the value of these benefits to workers, and the price of equivalent benefits purchased by individuals.

In this case, the best measure of the direct cost would be the “normal cost” measure used by pension fund actuaries, which in the public sector is based on the expected return on fund assets. This is the best estimate of how much employers need to contribute today to pay for future benefits. The measure preferred by Richwine and Biggs, however, is the largest value: the cost to individuals—specifically, workers with 401(k)s—of funding a similar guaranteed benefit. Richwine and Biggs also interpret this as the total cost to employers, and by extension to taxpayers, including not just the direct cost of pension contributions but also the indirect cost of assuming financial risk.

The two measures would be the same if employers (notably public employers) were as risk-averse as other investors, and if there were no other indirect costs and benefits to consider, such as employee retention. However, neither of these conditions holds; public employers are properly less risk-averse than most investors (especially individual investors), and pensions promote employee retention.

The logical implication of Richwine and Biggs’ position is that public employers and taxpayers would be indifferent between current pension funding practices and investing in Treasury securities, even though this would triple the cost of pension benefits. Richwine and Biggs would have a stronger case for putting a high price tag on the indirect cost of guaranteeing benefits if volatility in pension fund investment returns translated into large swings in state and local taxes. But pension funds are designed to absorb financial market volatility (that is, to diversify across time, not just across assets), since in any given year benefit outlays are typically a small fraction of assets. This allows pension funds to ride out bull and bear markets, unlike individual 401(k) savers, who need to tap all their retirement funds over a specified time period.

In the real world, when public employers face increases in pension costs large enough to warrant taxpayer concern, it is almost always because elected officials have neglected pension contributions, a problem that using the risk-free discount rate does not address. Even including funds to which elected officials neglected to make required pension contributions, as some did during the stock market bubble, a study by the Center for Retirement Research notes that contributions will need to rise by less than a third (from 3.8 percent to 5.0 percent of total state and government spending) to amortize the unfunded liabilities resulting from the bursting of the bubble if pension obligations are discounted using an expected return of 8 percent (Munnell et al. 2010). Though this is a significant increase in the wake of a severe downturn, it does not appear to justify tripling pension contributions to reduce similar risks in the future, especially considering that a significant share of current unfunded liabilities is due to underpayment as opposed to market volatility.

Furthermore, some of the risk of public pension funding falls on teachers and other workers, who typically pay for a portion of their benefits out of their paychecks. Though employee contributions are generally fixed in the short run, they often rise in the event of significant underfunding. In 2010–11, public employees in 18 states saw increases in employee contributions, not including increases that only affected new hires (Snell 2011). In addition, other forms of compensation may be cut back, such as salaries (which factor into pension benefits).3

In addition, Richwine and Biggs do not consider whether the indirect benefits to taxpayers of teacher pensions, such as teacher retention, may offset (or more than offset) these indirect costs. Traditional pension benefit structures inhibit mobility, since teachers who move frequently will tend to receive lower pensions than those who remain within one school district. This is a plus for employers and a minus for workers, who may pay a penalty if they want to change jobs. Conversely, the fact that employers assume financial and longevity risks associated with saving and investing for retirement is a plus for workers and a minus for employers. But large employers with long-term investment horizons, especially government entities, are much better equipped to assume these risks than individuals.

Retiree health benefits

While it is difficult to assign a value to future pension benefits, it is even more difficult to gauge the value of health benefits for future retirees.

Richwine and Biggs correctly point out that retiree health benefits are not included in the NCS. Based on a small sample of plans, they estimate the cost of these benefits at 8 percent of pay, comparable to the 7.6 percent estimate based on a much larger sample of public-sector workers by Munnell et al. (2011).

Munnell et al. point out that many employers are cutting back on retiree health benefits. Richwine and Biggs ignore not just the likelihood that some retiree health benefits will be cut, but also the uncertainty of these cutbacks, which, if Richwine and Biggs were consistent in their treatment of risk, would itself impose a cost on teachers. In addition, Keefe (2011) notes that these benefits are sometimes paid for through pension contributions, so there is the possibility of double counting. Rather than adjusting their estimate downward, however, Richwine and Biggs inappropriately inflate it to 10 percent of pay based on the fact that these benefits would be more costly to purchase in the individual market. This is a grossly flawed measure, as discussed earlier.

Though retiree health is the one area where the NCS does understate the cost of employee benefits, especially for teachers and other public-sector workers, Munnell et al. (2011) and Keefe (2010) find that public-sector workers are paid less even when taking these benefits into account. Admittedly, the problem of how to estimate the cost of future retiree health benefits is a difficult one, both because future health care costs are unknown and because these benefits may be cut back or eliminated at any time. For this reason, employers were not obliged to account for these future benefits as a liability on their balance sheets until recently.

“Work-year leave”

Just as the cost of retiree health benefits may be double counted in some areas, Richwine and Biggs tack on a “benefit”—time off for seasonal breaks, which they call “work-year leave”—that researchers usually incorporate into wage and salary comparisons. They value this benefit at 28.8 percent of wages, even though the NCS takes this leave into account in determining hourly wages and benefits.

Richwine and Biggs claim to have stumbled upon the issue of work-year leave in a footnote, even though the pertinent information is right in the body of the short Bureau of Labor Statistics article they cite (Schumann 2008). Furthermore, the same article cautions that teachers’ hours in the NCS are understated because the survey does not take into account the considerable time many teachers spend on lesson planning and grading at home, a point that indicates the NCS overstates teachers’ hourly compensation—and one that Richwine and Biggs ignore.

Richwine and Biggs’ rationale for including seasonal leave as a benefit rather than factoring the shorter work year into their salary comparison is that CPS earnings data may or may not account for teachers’ shorter work year, “so in many cases…weekly salaries in the CPS are simply annual salaries divided by 52 weeks.” Richwine and Biggs say that “[u]sing weekly salaries without further adjustment for summer vacation will upwardly bias teacher compensation.” In fact, it would downwardly bias teacher pay, but this is presumably an editing error.

The normal solution in this case would be to adjust CPS earnings measures, as necessary, to take into account teachers’ seasonal leave, since these adjustments have already been made to the NCS data that Richwine and Biggs use to compare benefits. Instead, Richwine and Biggs appear to include the unadjusted annual pay in their CPS “wage regression” results; they then tack on “work-year” leave as an additional benefit to the adjusted NCS data. They do not explain why they do this, though they imply it is because the CPS earnings data are unreliable (that is, the shorter work year is not consistently taken into account). If so, this is a problem for their overall analysis.

More likely, their unorthodox approach is designed to inflate teacher pay in comparison to that of private-sector workers, especially since Richwine and Biggs gloss over the rather startling implication of their results, which is that teachers appear to receive higher salaries than comparable private-sector workers, even without matching for “cognitive ability” (the exercise that takes up much of the first half of their paper and will be addressed in a forthcoming paper by Mishel and Roy). That is, they find that teachers are paid 80.7 percent as much as full-year private-sector workers even though they work only 71.2 percent of the year.

Compensating differentials and job security

Richwine and Biggs’ accounting for differences in working conditions appears equally arbitrary, especially when it comes to assigning a dollar value to job security.

In a perfectly competitive labor market, any pay gap between similarly skilled workers can be explained by what economists refer to as “compensating differentials”—differences in working conditions, job satisfaction, and the like. In practice, labor markets are far from perfectly competitive, and research often turns up results that seem to contradict this theory (e.g., many dangerous jobs, such as working in a meatpacking plant, pay poorly).

Richwine and Biggs treat job security as a form of compensation, akin to a fringe benefit, though many economists would treat it as a compensating differential. There are many reasons why turnover might be lower in some jobs than others, and only some of these could possibly be considered as equivalent to an employee benefit.

Low turnover is often viewed as a boon to both workers and employers, though there may be exceptions—for example, if low turnover reflects the difficulty of firing low-performing workers, or if pension and other benefits serve as “golden handcuffs” for workers. However, to the extent that low turnover reflects job satisfaction and a good employer/employee relationship, it is a win-win for workers and employers.

In the case of teachers, a large body of research finds that employee retention is very valuable to schools because teachers with at least three to five years of experience are much more effective than less experienced teachers (Boivie 2011). The longer a teacher stays within a school system, the easier it is for the employer to recoup the sunk costs of on-the-job training. This fact is not lost on school systems, as pensions and pay scales are designed to promote teacher retention through their years of peak effectiveness. Since reducing turnover is an explicit goal of teacher pensions, teacher retention should be counted against the cost of teacher benefits, not added to them. At the very least, it should not be considered an added cost to taxpayers.

In any case, Richwine and Biggs do not come close to proving that teachers have more job security than equally skilled private-sector workers, especially given recent mass layoffs in many school districts. Though Richwine and Biggs cite anecdotal evidence about incompetent teachers who manage to keep their jobs, there are also incompetent workers in the private sector, and Richwine and Biggs offer no evidence that incompetence is more tolerated in the public sector.

Richwine and Biggs also compare the drop in public education employment to the overall decline in private-sector employment in the recent downturn, but this is not a valid comparison since teachers should be compared to similarly skilled workers, not the entire private-sector workforce. Though it is possible that teachers’ employment is less cyclical than employment in other sectors, Munnell et al. (2011) show that public-sector workers have generally seen job losses similar to those of comparable private-sector workers in the recent downturn, after taking into account differences in education.

Finally, Richwine and Biggs compare unemployment rates from 2005–10 for occupations comparable to teaching. This is more pertinent than the comparison with all private-sector workers, though it still does not prove that teachers have more job security, or even job stability, than comparable workers. Keefe (2011) points out that differences in unemployment rates reflect not just the probability of job loss, but also the probability of new entrants obtaining a job in the first place. Thus, the low unemployment rate among teachers likely reflects, in part, that the supply of aspiring workers is lower in the teaching profession than in other professions due to teachers’ lower pay. Keefe also points out that to put a dollar value on job security, as Richwine and Biggs do, you would normally start by showing that people are willing to accept lower pay in exchange for a decreased likelihood of being laid off, though Keefe finds no empirical support for this compensating differential across occupations.

Even if teachers enjoyed more job security than comparable workers and were willing to forgo some pay in exchange, this begs the question of why Richwine and Biggs only attempt to put a monetary value on this single job characteristic, especially since they allude to others in the paper. In discussing private school teachers, for example, Richwine and Biggs acknowledge that the lower pay of some private school teachers might reflect the fact that “teachers in sectarian schools often consider their work to be part of their religious service, meaning they may accept below-market salaries.” They also note that “elite private schools often feature specialized curriculums directed at select groups of students.” In other words, many teachers derive personal satisfaction from their jobs and in working for the greater good (Almeida and Boivie 2009). Some teachers may also prefer to teach elite students or students of the same religion. However, other college-educated workers enjoy “perks” not factored into this analysis, such as more-flexible schedules.

As with the possibility that government employers get more “bang for the buck” from their benefits, the possibility that job satisfaction is greater for teachers than for similarly skilled workers may help explain the lower pay of teachers (and the even lower pay of some private school teachers) as well as their lower turnover. However, it is misleading to describe teachers as “overpaid” if their observed compensation is lower than that of other professionals. Furthermore, it would be a mistake to assume that school systems could further reduce teacher salaries or benefits relative to those of comparable occupations without affecting recruitment and retention.

Conclusion

Richwine and Biggs’ argument that teachers are overpaid because their benefits are twice as generous as those received by comparable private-sector workers is not persuasive. CPS and NCS data show that teachers earn significantly less in wage and salary compensation than comparable private-sector workers or those employed in large establishments, taking into account summer breaks and other differences in time spent at work. Meanwhile, NCS data show that they receive similar benefits to large-establishment workers, even though teachers are likely to be much better educated, on average.

The NCS data does not include the cost of retiree health benefits, which is hard to project with any degree of confidence. Whether or not retiree health benefits close the pay gap, the authors certainly do not prove that teachers are overpaid, let alone overpaid by half. Even if indirect costs and benefits are taken into account, Richwine and Biggs are highly selective in which of these costs and benefits to include, and improperly conflate the cost to individuals of purchasing similar benefits with the generally much lower cost to employers.

In particular, Richwine and Biggs triple the cost of teacher pensions by using a risk-free rate to value pension benefits, which they equate with the cost to individual 401(k) investors of funding equivalent benefits. While employers assume financial risks with defined benefit pensions, Richwine and Biggs do not take into account other indirect costs and benefits of these pensions, such as employee retention. There is no reason to believe that pensions’ important role in encouraging employee retention is more than offset by the financial risks employers assume with these plans.

The difference between the cost to employers and the value to workers of some benefits may help explain why public-sector workers appear willing to work for less pay. However, it does not mean that taxpayers are being “overcharged” for these benefits. In practice, the direct cost to employers is the only practical way to compare public-sector and private-sector pay because indirect costs and benefits are innumerable and impossible to measure directly. The direct cost of employee compensation is lower in the public sector, even according to Richwine and Biggs’ initial estimation. In short, their revised estimates are simply not convincing.

Endnotes

1. There are two measures of employer size: Firm size is the total number of employees in an organization, whereas establishment size is the number at a particular location. Unfortunately, CPS and NCS use different measures, though the two are obviously related. Though educational attainment is not readily available by establishment size (as opposed to firm size), it is likely that private-sector workers employed in large establishments (Richwine and Biggs’ comparison group) have more in common with private-sector workers employed by large firms than they do with public school teachers.

2. Teachers’ compensation as a percent of salary: 108.9 percent x (100 percent + 100.8 percent) = 218.7 percent

Private-sector compensation as a percent of salary: (100 percent + 43.5 percent) = 143.5 percent

Teachers’ pay premium: (218.7 percent – 143.5 percent) / 143.5 percent ≈ 52 percent.

3. Additional problems with using the risk-free rate to discount projected pension obligations are discussed in Baker (2011a; 2011b), Lav and McNichol (2011), and Morrissey (2011). Gollier (2007) discusses the role of pension funds in intergenerational risk sharing, which allows employers and taxpayers to take advantage of the equity premium.

References

Allegretto, Sylvia A., Sean P. Corcoran, and Lawrence Mishel. 2004. How Does Teacher Pay Compare? Methodological Challenges and Answers. Washington, D.C.: EPI. http://www.epi.org/publication/books_teacher_pay/

Allegretto, Sylvia A., Sean P. Corcoran, and Lawrence Mishel. 2008. The Teaching Penalty: Teacher Pay Losing Ground. Washington, D.C.: EPI. http://www.epi.org/publication/book_teaching_penalty/

Allegretto, Sylvia A., Sean P. Corcoran, and Lawrence Mishel. 2011. The teaching penalty—an update through 2010. Economic Policy Institute Issue Brief #298. Washington D.C.: EPI. http://www.epi.org/publication/the_teaching_penalty_an_update_through_2010/

Almeida, Beth and Ilana Boivie. 2009. “The Staying Power of Pensions in the Public Sector,” CPER Journal, no. 195. http://www.nirsonline.org/storage/nirs/documents/staying_power_of_public_pensions.pdf

Baker, Dean. 2011a. The Origins and Severity of the Public Pension Crisis. Washington, D.C.: Center for Economic and Policy Research. http://www.cepr.net/index.php/blogs/cepr-blog/returns-on-public-pensions-what-rates-should-we-assume

Baker, Dean. 2011b. “Returns on public pensions: What rates should we assume?” CEPR Blog (blog), March 6. http://www.cepr.net/index.php/blogs/cepr-blog/returns-on-public-pensions-what-rates-should-we-assume

Bender, Keith A. and John S. Heywood. 2010. Comparing Public and Private Sector Compensation over 20 Years. Milwaukee: University of Wisconsin. http://www.slge.org/index.asp?Type=B_BASIC&SEC=%7B22748FDE-C3B8-4E10-83D0-959386E5C1A4%7D&DE=%7BBD1EB9E6-79DA-42C7-A47E-5D4FA1280C0B%7D

Biggs, Andrew G. 2002. The stock market and Social Security reform. Washington, D.C.: Cato Institute. http://www.cato.org/pub_display.php?pub_id=3562

Biggs, Andrew G. and Jason Richwine. 2011. Public vs. Private Sector Compensation in Ohio: public workers make 43 percent more in total compensation than their private-sector colleagues. Columbus: Ohio Business Roundtable. http://media.cleveland.com/open_impact/other/Ohio%20Business%20Rountable%20study.pdf

Boivie, Ilana. 2011. The three Rs of teacher pension plans: recruitment, retention and retirement. Washington, D.C.: National Institute on Retirement Security. http://www.sdcera.org/PDF/Teachers_Pension_Plan_Study_Oct_2011.pdf

Gollier, Christian. 2007. “Intergenerational Risk-Sharing and Risk-Taking of a Pension Fund.” Journal of Public Economics, vol. 92, no. 5–6, pp. 1463–1485.

Hanauer, Amy. 2011. Business Roundtable Study Deeply Flawed. Cleveland: Policy Matters Ohio. http://www.policymattersohio.org/wp-content/uploads/2011/10/BRT-Study_2011.pdf

Keefe, Jeffrey H. 2010. Debunking the myth of over-compensated public employees. Economic Policy Institute Briefing Paper #276. Washington, D.C.: EPI. http://www.epi.org/publication/debunking_the_myth_of_the_overcompensated_public_employee/

Keefe, Jeffrey H. 2011. Ohio public employees are not overcompensated: Rebutting a diversion from Senate Bill 5. Economic Policy Institute Briefing Paper #329. Washington, D.C.: EPI. http://www.epi.org/publication/ohio-public-employees-overcompensated-senate-bill-5/

Keefe, Jeffrey H. 2012. “Review of assessing the compensation of public-school teachers.” Think Tank Review. Boulder, Colo.: National Education Policy Center. http://nepc.colorado.edu/files/TTR-TchrCompens-Heritage_0.pdf

King, Miriam, Steven Ruggles, J. Trent Alexander, Sarah Flood, Katie Genadek, Matthew B. Schroeder, Brandon Trampe, and Rebecca Vick. 2010. Integrated Public Use Microdata Series, Current Population Survey: Version 3.0. [Machine-readable database]. Minneapolis: University of Minnesota.

Lav, Iris J. and Elizabeth McNichol. 2011. Misunderstandings regarding state debt, pensions, and retiree health costs create unnecessary alarm. Misconceptions also divert attention from needed structural reforms. Washington D.C.: Center on Budget and Policy Priorities.  http://www.cbpp.org/files/1-20-11sfp.pdf

Morrissey, Monique. 2011. Discounting public pensions: Reports of trillions in shortfalls ignore expected returns on assets. Economic Policy Institute Policy Memorandum #179. Washington, D.C.: EPI. http://www.epi.org/publication/pm179/

Munnell, Alicia H., Jean-Pierre Aubry, Josh Hurwitz, and Laura Quinby. 2011. Comparing compensation: State-local versus private sector workers. Boston: Center for Retirement Research. http://crr.bc.edu/images/stories/Briefs/slp_20.pdf

Richwine, Jason and Andrew G. Biggs. 2011. Assessing the compensation of public-school teachers. A Report of the Heritage Center for Data Analysis. CDA 11-03. Washington, D.C.: The Heritage Foundation and American Enterprise Institute. http://www.aei.org/files/2011/11/02/-assessing-the-compensation-of-publicschool-teachers_19282337242.pdf

Schmitt, John. 2010. The wage penalty for state and local government employees. Washington, D.C.: Center for Economic and Policy Research.  http://www.cepr.net/documents/publications/wage-penalty-2010-05.pdf

Schumann, Richard. 2008. Work schedules in the national compensation survey. Washington, D.C.: Bureau of Labor Statistics. http://www.bls.gov/opub/cwc/cm20080722ar01p1.htm

Snell, Ron. 2011. “State Retirement Legislation in 2010 and 2011.” National Conference of State Legislatures, PowerPoint presentation presented June 30, 2011. http://www.ncsl.org/LinkClick.aspx?fileticket=WCg6SYg6vZ4%3d&tabid=13399

According to the data, they’re fundamentally right. This paper presents 12 figures that demonstrate how skewed economic rewards (in income, wages, capital income, and wealth) have become in the United States. These figures, most of which cover 1979 through 2007 (prior to the recession) generally break out trends for the top 1 percent, the next richest 9 percent, and then the bottom 90 percent of households or earners. While income growth at the very top—the richest 1 percent and above—has been truly staggering, incomes at roughly the 90th percentile and above (the richest 10 percent) have generally at least matched the rate of economy-wide productivity. It is below the 90th percentile where one really sees the potential fruits of economic growth (as measured by economy-wide productivity) failing to reach American households. An economy that fails to cut in 90 percent of American households on a fair share of economic growth is one that needs serious reform. As the figures show:

• The top 1 percent of households have secured a very large share of all of the gains in income—59.9 percent of the gains from 1979–2007, while the top 0.1 percent seized an even more disproportionate share—36 percent. In comparison, only 8.6 percent of income gains have gone to the bottom 90 percent. The patterns are similar for wages and capital income.
• As they have accrued a large share of income gains, the incomes of the top 1 percent of households have pulled far away from the incomes of typical Americans. In 2007, average annual incomes of the top 1 percent of households were 42 times greater than incomes of the bottom 90 percent (up from 14 times greater in 1979) and incomes of the top 0.1 percent were 220 times greater (up from 47 times greater in 1979).
• The financial sector’s share of the overall economy has roughly doubled in recent decades, and now stands at 7.6 percent of total national income. Relative to this sector’s share in 1979, this translates into an extra $547 billion in compensation and profits claimed by the sector—a trend with questionable social payoff.
• Growth in wealth, not just incomes, has also become greatly skewed in recent decades. Most of the wealth gains of the last generation went to those who already had the most wealth, a group increasingly distant from the vast American middle-class. The wealth of the median household actually declined over this time period. As a result, in 2009, wealth held by the wealthiest 1 percent of households was 225 times greater than that held by the median household.

The effect of policy on income and wealth inequality

No one who has looked at trends in economic inequality in the United States in recent decades could dispute the dramatic increase in the share of all income claimed by the richest subgroups—especially the highest-earning 1 percent referred to by Occupy Wall Street activists when they say they represent the 99 percent of Americans left behind. Mishel, Bernstein, and Shierholz (2009) present a comprehensive review of these trends and Piketty and Saez (2010, updating earlier reports) explore in more depth the gains enjoyed by the top 1 percent.

Quick Facts: The Upside-Down Economy

Download and print half-sheet fliers that provide easy-to-understand information about the widening gap in economic status between the bottom 99 percent and the top 1 percent of Americans. The fronts provide numerical facts about the subject area while the backs contain definitions, explanations, and an elaboration of each issue:

Income

Wages

Wealth

 

 

There is some disagreement around the edges of the debate concerning just how dramatic this income-share increase was or when exactly it happened—was it steady and continuous, or the result of a couple of discrete “jumps”? And there are those who discount the seriousness of the divide, saying that middle-class incomes are managing to grow despite the huge increase in the top earners’ share. But no serious analyst denies that the top 1 percent (of households or tax-units or families) has seen a very large increase in incomes and in share of total income since the late 1970s.

Public policy, either through commission or omission, has played a central role in the increasing concentration of income. For example, Baker (2006), Bivens (2010), and Hacker and Pierson (2010) have all documented the role of various policies in generating greater inequality. The decade-long surge in income inequality occurred in pre-tax incomes, driven by developments in both major kinds of market-based incomes, namely the wage and salary incomes from work, and capital incomes (realized capital gains, interest, dividends) from wealth. And we know that the most obvious way policy can affect incomes—through taxes—has clearly aided the widening of the income gap. The Congressional Budget Office (CBO) shows that even as their share of total incomes more than doubled between 1979 and 2007, the richest 1 percent of household’s effective federal tax rate fell from 37 percent to 29.5 percent.

The clear policy tilt in favor of the highest-income households in the completely visible realm of taxes suggests that this group receives preferential treatment in the much more opaque policy decisions that get made in Washington every day. For example, Bartels (2007) shows how policymakers give much larger weight to the preferences of richer constituents.

What the Occupy Wall Street movement has done with its “We are the 99 percent” campaign is to remind Americans that economic outcomes are not just like the weather, something that must simply be endured and adapted to rather than forced to change. Instead, economic outcomes are shaped by political decisions. This insight is valuable because it confers the power to challenge the status quo, which is often preserved by claims that economic rewards are doled out through simple meritocracy and that any interference with market outcomes will wreck the economy. It’s not so. Markets are always shaped by policy, and policies in the United States have been shaped to benefit the already well-off. Changing the rules to ensure that rewards are more broadly shared can lead to an economy that is both more efficient and more fair.

The widening income gap

Figures A–C display trends in growth of overall market incomes, including wages and salaries as well as interest, dividend, and capital income generated by holding wealth. In the long period before the current recession, from 1979 to 2007, inflation-adjusted average annual incomes of the highest-income 1 percent of households grew by 224 percent, as shown in Figure A. Those even better off, the top 0.1 percent (the highest-income one one-thousandth of households), saw their incomes grow by 390 percent. In contrast, incomes of the bottom 90 percent grew just 5 percent between 1979 and 2007—and all of that growth occurred in the unusually strong income growth that occurred from 1997 to 2000, a period followed by declining income from 2000 to 2007.¹ These data include all sources of market-based incomes such as wages and salaries, dividend and interest income, and realized capital gains, but do not include government transfer income (such as Social Security income or unemployment benefits).

Figure A

Because of their vastly greater income growth, the highest-earning 1 percent of households have rapidly distanced themselves from the vast majority (the bottom 90 percent). As Figure B shows, average annual incomes of the top 1 percent of households in 1979 were 14 times greater than incomes of the bottom 90 percent; by 2007 incomes of the top 1 percent were 42 times greater. The income gap between the upper 0.1 percent of households and the bottom 90 percent grew even more, from a top-to-bottom ratio of 47-to-1 in 1979 to 220-to-1 in 2007.

Figure B

The vastly greater income growth of the top highest-income 1 percent of households also obtained a much larger share of income growth than the vast majority (the bottom 90 percent). As shown in Figure C, the top 1 percent gained 59.9 percent of all the income growth generated between 1979 and 2007. In contrast, the bottom 90 percent received just 8.6 percent of all the income generated over the same period. It’s illuminating to note that the bottom 90 percent were able to claim just one-fourth of what the top one one-thousandth of households claimed from the growth of that period (36 percent).

Figure C

Rising inequality in income from work

Figures D–F examine the rising inequality of wage and salary income—in other words, income from work. Labor earnings are by far the most evenly distributed sources of overall income because, after all, the vast majority of non-retired households have members that work. Yet labor earnings have become much more unequally distributed in recent decades. Figure D shows that the top 1 percent of wage and salary earners increased their inflation-adjusted average annual salaries by 144% from 1979 to 2006. The top one one-thousandth (0.1 percent) of earners enjoyed annual wages growth of 324 percent over that same period.

Figure D

In contrast, the bottom 90 percent of wage earners increased their annual salaries by about 15 percent from 1979 to 2006. Most of this growth occurred during the relatively brief period of tight labor markets that accompanied the late 1990s boom. Between 1979 and 1995, average annual wages for the lowest-earning 90 percent grew just 2.8 percent. And from 2000 and 2006, wages did not improve at all. Thus, nearly all of the wage and salary growth of the bottom 90 percent from 1979 to 2006 occurred from 1995 to 2000 when unemployment was falling and then remained low.²

As with overall incomes, the disparity in wage growth has significantly widened the gap in salary levels between the top earners and everyone else, as shown in Figure E. In 1979 average annual salaries of the top 1 percent of wage earners were 9.4 times that of those in the bottom 90 percent, but by 2000 the gap had more than doubled to 20-to-1, a level that was maintained until 2006.

Figure E

The very highest-wage earners—those in the upper 0.1 percent (the top one one-thousandth)—increased their distance from the earners in the bottom 90 percent even more rapidly; the ratio of their earnings to those in the bottom 90 percent rose from 21-to-1 in 1979 to 80-to-1 in 2000. This gap shrank after the stock market bubble burst in the late 1990s (wage data include the “realized stock options” that top corporate officers receive) but had nearly recovered its former size by 2006.

Figure F looks directly at the ratio of average compensation earned by the chief executive officers of large firms relative to the compensation of typical workers. In 1978, CEO compensation was 35 times greater than that of the typical worker, up from 24 times as great in 1965. After 1979 the pay of CEOs skyrocketed; by 2000 their pay was 299 times that the pay of a typical worker.

Figure F

That level of CEO pay was admittedly somewhat inflated by the stock market boom in the late 1990s, and retreated significantly after the tech bubble burst. However, by 2007, CEO pay had nearly restored itself, attaining a ratio of 277-to-1 relative to pay of a typical worker. CEO pay fell again relative to typical workers in the Great Recession but is again reestablishing itself in the recovery. In 2010, the ratio of 243-to-1 was the fifth highest of any year since 1965. At this rate, it will likely not take long for the gap to reach its prior peak.³

Increasing concentration of income from wealth-holding

Figures G–H show that the trend of rapidly growing concentration in overall income and labor earnings is also apparent in the growth of income earned from wealth-holding, often labeled either “unearned” or “capital” income. Essentially, capital incomes are always and everywhere less equally distributed than wage income. As shown in Figure G, in 1979 the top 1 percent of households on the income scale already claimed 38 percent of all capital income generated in the economy. By 2007 this share had ballooned to 57 percent. The next richest 9 percent saw their share of capital incomes shrink from 29 percent in 1979 to 23 percent in 2007. And the bottom 90 percent, which collected 33 percent of capital incomes in 1979, claimed only 20 percent by 2007. This startling concentration of already unequally distributed capital incomes defies the logic of claims that there is a natural limit to how much of the fruits of economic growth can go to any one group.

Figure G

The very large rise in the share of all capital incomes collected by the highest-income 1 percent since 1979 means that this group has also collected a disproportionate share of the growth in these incomes over the same period. Basically, if the top 1 percent’s share of all capital incomes had remained constant between 1979 and 2007, they would have claimed 37 percent of capital income growth in the economy in those years. Instead, as Figure H shows, the top 1 percent alone collected a whopping 86.5 percent of growth in capital incomes during this period.⁴ The next highest-income 4 percent claimed 10.7 percent of all capital income growth while the bottom 95 percent claimed just 2.8 percent of the growth in these incomes. This figure departs from the convention of the other charts in not isolating the bottom 90 percent because their average capital incomes fell between 1979 and 2007, registering as negative capital income growth, which is hard to depict in a pie chart.

Figure H

The financial sector’s increasing claim on growth

Much of the rising share of total income claimed by the top-earning 1 percent is associated with the rise of the financial sector, which is a dominant employer at the top. Figure I shows the share of total gross domestic product, or national income, attributable to compensation and profits in the corporate financial sector. Between 1929—just before the Great Depression ended the first Gilded Age—and 1973, this share fell from 3.7 percent to 3.2 percent. But between 1973 and 2007, this share more than doubled, to nearly 7 percent.

Figure I

And financial sector compensation and profits’ share of GDP rebounded quickly from the dip of the Great Recession and actually passed its pre-recession peak. By 2010, in fact, the rising share of finance translates into an extra $547 billion claimed by this sector relative to the case where its share had remained at its 1979 level (3.8 percent). This is serious money. The payoff to these larger claims made by the financial sector are dubious. For example, business investment in plant and equipment (i.e., the productivity-generating investment that financial firms are supposed to make cheaper and safer) did not rise between 1973 and 2007. Residential investment, outside of the bubble-inflated mid-2000s, has also failed to show any persistent upward climb during the time that the financial sector has claimed an ever-larger piece of the pie. It is, in short, not off-base to wonder whether there is any return to forking over a much larger share of economic activity to the financial sector.

The concentration of wealth

The concentration of wealth has mirrored trends in the concentration of income. Wealth is a measure of a household’s assets (such as real estate, stocks, bonds, and cash) minus their liabilities (such as home mortgages and other personal debt). The only available data covering recent decades dates back to 1983 and shows that the wealth held by the wealthiest 1% of households grew far more than the wealth of the median household, whose wealth was actually lower in 2009 than in 1983. Figure J shows that the wealth of the top 1 percent grew over the 1980s and ‘90s and by 2007 was 103 percent greater than in 1983. The financial crisis in 2008 reduced the wealth of those at the top but by 2009 their wealth remained 48 percent greater than in 1983. The median household’s wealth fared far worse. After falling in the early 1990s the median household’s wealth rose and was 48 percent greater in 2007 than in 1983. But the fall of wealth in the financial crisis was sharper for those in the middle than at the top because those in the middle have much of their wealth in housing, values of which fell dramatically after the housing bubble burst. By 2009 the median household’s wealth had fallen so much that their wealth was 13.5 percent less than what it was in 1983.⁵

Figure J

Not surprisingly, the gap between the wealth of those at the top and those in the middle substantially grew over the last few decades, as Figure K shows. In 1983 the wealthiest 1 percent of households had wealth that was 131 times greater than wealth of the median household. This gap grew until the early 1990s and again in the 2000s, and by 2009 the top 1 percent had 225 times as much wealth as the median household.⁶

Figure K

Perhaps more startlingly, more than 94 percent of the gains in wealth from 1983 to 2009 accrued to the top fifth of wealthiest households, with 40.2 percent of the gains going to the wealthiest 1 percent and 41.5 percent going to the next wealthiest 4 percent of households (Figure L). This translated to gains among the wealthiest 1 percent of $4.5 million per household and gains among the next wealthiest 4 percent of roughly $1.2 million per household.⁷

Figure L

In other words, the richest 5 percent of households obtained roughly 82 percent of all the nation’s gains in wealth between 1983 and 2009. The bottom 60 percent of households actually had less wealth in 2009 than in 1983, meaning they did not participate at all in the growth of wealth over this period.

Basing policy in the true picture of income and wealth

The insights offered by the data on income, wealth, and inequality should shape the economic policy debate going forward. Most immediately, they should inform budget deficit debates about what the United States can “afford.” The nation can easily afford more federal government support aimed at reducing today’s historically high and persistent rates of joblessness. In fact it is the cheapest option in all major economic respects (Mishel 2011).

Once the current crisis of joblessness has passed and smaller imbalances between federal investment and revenues are appropriately targeted, attention should turn to supporting the same level of economic security and dignity that we have provided for generations. This would mean ending the unnecessary calls to close budget deficits by cutting the benefits provided by Social Security, Medicare, and Medicaid.

Thirty years of economic data show that the U.S. economy has generated significant levels of income and should continue to do so into the future; in other words, there is no economic constraint that mandates that we scale back expectations for living standards growth in coming years (Mishel 2011). But this vast income that has been generated has been distributed in an extremely skewed fashion; typical American families have not benefitted from it nearly as much as they could have. This is a political problem that, if solved, has the potential to make our country more fair and the vast majority of its citizens more prosperous.

The politics of economic policymaking may be broken, but the U.S. economy is not broke, the data show. The country does have the economic wherewithal to provide a decent standard of living for all.

Endnotes

1. Economic Policy Institute analysis of “Table A6: Top fractiles income levels (including capital gains) in the United States” from “Income Inequality in the United States, 1913-1998” with Thomas Piketty, Quarterly Journal of Economics, 118(1), 2003, 1-39 (Longer updated version published in A.B. Atkinson and T. Piketty eds., Oxford University Press, 2007) (Tables and Figures updated to 2008 in Excel format, July 2010).

2. Based on Table 3.10 in Mishel, Bernstein, and Shierholz (2009), which uses data from Kopczuk, Saez and Song (2007), Table A-3. Data in Table 3.10 for 2006 was extrapolated from 2004 data using growth rates from Social Security Administration wage statistics (http://www.ssa.gov/OACT/COLA/awidevelop.html). SSA provides data on share of total wages and employment in annual wage brackets such as for those earning between $95,000.00 and $99,999.99. We employ the midpoint of the bracket to compute total wage income in each bracket and sum all brackets. Our estimate of total wage income was 99.1 percent of the actual. We used interpolation to derive cutoffs building from the bottom up to obtain the 0–90 percent bracket and then estimating the remaining categories. This allowed us to estimate the wage shares for upper wage groups. To obtain absolute wage trends we used the SSA data on the total wage pool and employment and computed the real wage per worker (based on their share of wages and employment) in the different groups.

3. The CEO pay data are described in the table note for table 3.41 in Mishel, Bernstein, and Shierholz (2009).The compensation data for typical workers comes from the Bureau of Labor Statistics’ series on average hourly earnings of production, non-supervisory workers inflated to compensation using the ratio of compensation to wages in the Bureau of Economic Analysis National Income and Product Accounts.

4. The data in Figure G comes directly from the Congressional Budget Office, which calculates the share of all capital income going to various income groupings. Figure H is calculated by EPI with slightly different data, specifically the CBO estimates of average incomes’ sources of incomes by income groupings. What are being labeled as growth in capital incomes between 1979 and 2007 in Figure H are dividends, interest payments, capital gains, and “other business income,” which includes partnership income, income from S corporations, and rental income.

5. The data on wealth are based on Wolff’s analysis of the Federal Reserve Board’s Survey of Consumer Finances presented in Table 3 of Allegretto (2010)

6. Ibid.

7. Ibid.

References

Allegretto, Sylvia. 2009. The State of Working America’s Wealth, 2011: Through Volatility and Turmoil the Gap Widens. Economic Policy Institute Briefing Paper #292. Washington, D.C.: EPI.

Bartels, Larry M. 2008. Unequal Democracy: The Political Economy of the New Gilded Age. Princeton, N.J.: Princeton University Press.

Bivens, Josh. 2011. Failure by Design: The Story behind America’s Broken Economy. An Economic Policy Institute Book. Ithaca, N.Y.: ILR Press, an imprint of Cornell University Press.

Congressional Budget Office (CBO). June 2010. “Average Federal Tax Rates for All Households, by Comprehensive Household Income Quintile.” Washington, D.C.: CBO. http://www.cbo.gov/publications/collections/tax/2010/all_tables.pdf

Hacker, Jacob S. and Paul Pierson. 2010. Winner-Take-All Politics: How Washington Made the Rich Richer – And Turned Its Back on the Middle Class. New York: Simon & Schuster.

Mishel, Lawrence. 2011. We’re not broke nor will we be: Policy choices will determine whether rising national income leads to a prosperous middle class. Economic Policy Institute Briefing Paper #310. Washington, D.C.: EPI. http://www.epi.org/publication/were_not_broke_nor_will_we_be/

Mishel, Lawrence, Jared Bernstein, and Heidi Shierholz. 2009. The State of Working America 2008/2009. An Economic Policy Institute Book. Ithaca, N.Y.: ILR Press, an imprint of Cornell University Press.

Piketty, Thomas and Emmanuel Saez. 2010. Excel tables and figures with 2008 data updating “Income Inequality in the United States, 1913–1998,” Quarterly Journal of Economics, 118(1), 2003, 1–39 (longer updated version published in A.B. Atkinson and T. Piketty eds., Oxford University Press, 2007).

The Current Population Survey reports that the total number of unemployed workers in August was 14.0 million. The ratio of unemployed workers to job openings was thus 4.6-to-1 in August, a deterioration from the July ratio of 4.3-to-1. By comparison, in December 2000 the job seeker’s ratio was 1.1-to-1. Furthermore, the highest this ratio ever got in the early 2000s downturn was 2.8-to-1. August marks three years straight that the job seeker’s ratio has been at or above 3-to-1. Put another way: we’ve exceeded the highest level reached in the early 2000s recession for the last three years straight. And we’ve been substantially above 4-to-1 for the last two years and eight months.  A job seeker’s ratio of more than 4-to-1 means that for more than three out of four unemployed workers, there simply are no jobs. Two years and eight months—139 weeks—of a job seeker’s ratio above 4-to-1 is why the current extended unemployment insurance benefits, which last a maximum of 99 weeks, remain crucial. Furthermore, extending these benefits will not contribute to keeping the unemployment rate high, as some economists have claimed.

With research assistance from Nicholas Finio and Hilary Wething

The Teacher Gap

In September, public-sector employment dropped by 34,000, with most of that  (-33,000) occurring in state and local governments.  Over the last three years, state and local government employment has dropped by 641,000, as state and local budgets have been squeezed as a result of the recession. With kids heading back to the classroom this fall, it’s worth considering how much of that drop has hit public schools.

Almost half (-278,000) of the decline in state and local government jobs was in local public  education, which is largely jobs in public K-12 education (and the majority of workers  in public K-12 education are teachers, but there are also teacher aides, librarians, guidance counselors, administrators, support staff, etc.). On the other hand, over the same period, public K-12 enrollment increased by 0.6 percent (using the actual and projected enrollment growth rates found in Table 1 here). Just to keep up with this growth in the student population, employment in local public education should have grown at roughly the same rate, which would have meant adding around 48,000 jobs. Putting these numbers together (i.e., what was lost plus what should have been added to keep up with the expanding student population) means that the total jobs gap in local public education as a result of the Great Recession and its aftermath is around 326,000 jobs.

This decline means not only larger class sizes, but also fewer teacher aides, fewer extra-curricular activities, and a narrower curriculum for our children. Furthermore, this number almost surely understates the real gap. Between 2008 to 2010, the number of children living in poverty increased by 2.3 million, and is likely even higher today. Increased child poverty increases the need for services provided through schools. Instead, public schools have fewer personnel and fewer resources to educate more students, and more students with greater needs.

Underemployment

The “underemployment rate” (the U-6 measure of labor underutilization) is the BLS’s most comprehensive measure of labor market slack: it includes not just the officially unemployed and the marginally attached (jobless workers who want a job and are available to work but have given up actively seeking work), but also people who want full-time jobs but have had to settle for part-time work. This measure increased in September from 16.2% to 16.5%, due to a very large increase of  444,000 “involuntary” part-time workers. In September there were a total of 25.8 million workers who were either unemployed or underemployed (14.0 million officially unemployed, 2.6 million marginally attached, and 9.3 million involuntary part-time workers). Racial and ethnic minorities have been particularly hard hit by underemployment.

Long-term unemployment

The share of unemployed workers who have been unemployed for more than six months increased to 44.6% in September, not far off its record high of 45.6% in the Spring of 2010. The number of workers who have been unemployed for more than six months increased by 208,000 to a total of 6.2 million in September.  The number of unemployed who had been jobless more than a year was 4.4 million in September (not seasonally adjusted), up from 4.3 million one year ago.  This is unsurprising given that there have been more than four unemployed workers per job opening since January 2009.

Labor force participation and the employment-to-population ratio

The labor force participation rate (LFPR) increased to 64.2% in September, back to where it was in the Spring.  September’s LFPR increase came from workers under 25 and over 54, while the LFPR of “prime age” workers (workers age 25-54) held steady.  Remarkably, the overall labor force has seen almost no net growth (+81,000) since the recession started, though the working-age population grew by over seven million in that time. There are currently 2.6 million “marginally attached” workers in this country.  If these workers were in the labor force and counted as unemployed, the unemployment rate would be 10.5% (the U-5 measure of labor underutilization) instead of 9.1%.

At a time like this, with the labor force not growing at a steady pace, arguably the cleanest measure for assessing labor market trends is the employment-to-population ratio, which is simply the share of the working-age population that has a job. This measure increased by one-tenth of a percentage point to 58.3% in September.  Again, the increase occurred among workers under 25 and over 54, with workers age 25-54 seeing a decline from 75% to 74.9%, near the low for the downturn of 74.7% in December 2009.

Earnings

Hourly wages increased by 4 cents in September.  In the last three months, wages have grown at a 1.9% annualized rate, and they have grown 1.9% over the last year. This remains far below the pre-recession growth rate, as persistent high unemployment has put strong downward pressure on wage growth. Average weekly wages growth has been similar, growing at a 1.9% annualized rate over the last three months and 2.1% over the last year.

Demographic breakdowns

Unemployment in September was 9.7% for those with only a high school education, and 4.2% for those with a college degree or more.  While workers with higher levels of education have lower unemployment rates, all education categories have seen their unemployment rates roughly double over the last four years, running counter to the notion that a key part of today’s unemployment is due to employers being unable to fill their demand for skilled workers.

Considering additional breakdowns by age, race/ethnicity, and gender, we find that all major groups of workers have experienced substantial increases in unemployment over the Great Recession and its aftermath.  However, young workers, racial and ethnic minorities, and men have been hit particularly hard.

• In September, unemployment was 17.4% among workers age 16–24, 8.1% among workers age 25–54, and 6.7% among workers age 55 and older (up 5.7, 4.0, and 3.5 percentage points, respectively, since the start of the recession in December 2007).
• Among workers younger than age 25 who are not enrolled in school, unemployment over the last year averaged 21.6 % for those with a high school degree, and 9.6% for those with a college degree (reflecting increases of 9.6 and 4.2 percentage points, respectively, since 2007).
• Unemployment in September was 16.0% for African American workers, 11.3% for Hispanic workers, and 8.0% for white workers (up 7.0, 5.0, and 3.6 percentage points, respectively, since the start of the recession).
• Unemployment was 9.4% for men, compared with 8.7% for women (up 4.3 and 3.8 percentage points, respectively, since the start of the recession).

Temporary help services and average hours

One point of relatively good news in the report was that employment in temporary help services increased by 19,000 in September, and there were upward revisions to the prior two months’ data.   The average growth rate in temporary help services in the third quarter was 18,000, compared an average decline of 4,000 in the second quarter.  Since hiring in temporary help services tends to be a signal of broader upcoming hiring patterns, the fact that temporary help is not declining is good news.  However, it remains below the average growth rate of 28,000 from the fourth quarter of 2009 through the first quarter of 2011.

Furthermore, it is important to note that employment in temporary help services remains far below where it was before the recession started.  One thing this underscores is that the lack of hiring right now indicates a lack of demand, not concern on the part of businesses about regulatory burdens.  If businesses needed workers to meet demand but were reluctant to hire because of some other reason, we would see a big flood into temporary help services, and that is not happening.  While the temporary help services sector has recouped a greater share of its losses than other private sector employment, its losses during the recession were much more severe.  Employment in temporary help services is now 10.3% below where it was in December 2007, while employment in other private-sector employment is now 5.3% below where it was in December 2007.

A similar logic applies to average hours.  The length of the average workweek increased in September back to its July value of to 34.3 hours.  However, average hours have dropped since the spring, when it was at 34.4, and have thus far made up just two-thirds of what they lost in the first 18 months of the downturn (the low point was 33.7 hours in June 2009).  Again, if demand was there but businesses didn’t want to hire for some other reason, we would see them strongly ramping up the hours of the workers they have.  As it is, there remains substantial room to meet unmet demand by increasing hours of existing workers; if private-sector employers were to simply restore the hours of their workers back to pre-recession levels, it would be equivalent to adding over 900,000 jobs at current average hours.

Conclusion

The U.S. is currently 6.6 million jobs below where it was when the recession started. But because the working-age population grows as the population expands, in the three years and nine months since the recession started we needed to have added around 4.5 million jobs to keep the unemployment rate from rising. Putting these numbers together means the current gap in the labor market is roughly 11.1 million jobs. To fill that gap in three years—by Fall 2014—while still keeping up with the growth in the working-age population—would require adding around 400,000 jobs every single month. To fill the gap in five years—by Fall 2016—would mean adding 280,000 jobs each month. By comparison, over the last three months, the economy added just 96,000 jobs per month on average.  At this rate, the unemployment rate will not come down.  More than two years into the official recovery, the United States has yet to produce anything close to the rate of job growth that will put its backlog of unemployed workers back to work before the end of the decade.  The key issue holding back job growth is a lack of demand, not regulatory uncertainty or anything else. (For policies that we can and should be pursuing to stimulate demand and generate jobs, see EPI’s Briefing Paper, Putting America Back to Work.)

— Research assistance by Nick Finio, Natalie Sabadish, and Hilary Wething

Problems with Biggs and Richwine’s Analysis

• Double counting Ohio retiree health benefits—which in Ohio are prefunded from employer and employee pension contributions—by counting them in addition to employer pension-fund contributions.
• Treating Ohio retiree health benefits as guaranteed benefits, when they are discretionary benefits.
• Using a 4 percent interest rate to project the value of future pension assets, when the Ohio Public Employees Retirement System (OPERS) has achieved 8.99 percent rate of return over the last 30 years. This undervaluation of pension assets inflates public employee compensation costs by overestimating what employers must contribute to meet future pension obligations.
• Calculating future pension liabilities based upon wage increases as if they are fixed and guaranteed when instead they are negotiated and depend on many environmental factors. This overvaluation of pension liabilities adds to the overestimation of what employers must contribute to meet future pension obligations.
• Estimating employee pension contributions as if they are fixed, though in fact they can be increased by the plan trustees if necessary to compensate for any shortfalls arising from a lower than expected return on pension asset investments.
• Arbitrarily reducing private-sector employer tax contributions for Social Security from 6.2 percent of payroll (up to $106,200) to 2 percent of payroll because they do not believe the Social Security benefits are worth 6.2 percent.
• Arguing that public employee job security confers a premium worth 9.3 percent that should be added to the estimated cost of employee compensation, even though there is no guaranteed job security in the Ohio public sector and no empirical evidence offered by the authors to support their claim of a job security premium or penalty.
• Making a number of important calculation obfuscations, including discount rate adjustment factor and normal cost of pensions that could not be replicated.

Furthermore, Biggs and Richwine’s analysis focuses on issues that are outside the scope of collective bargaining in Ohio and thus not pertinent to the debate over SB 5. Because retiree health benefits, the guaranteed nature of public-sector pensions, and job security are not subjects of public sector collective bargaining, the parties cannot enter into a binding enforceable contract involving any of these issues. Rather, retiree health benefits and public-sector pensions are legislated, not bargained, and management retains the right to set employment levels in public employment and can hire, fire, and lay off employees.

As Biggs and Richwine observe, “The main impact of SB 5, however, would be to reduce the number of employment issues that are collectively bargained …. SB 5 would disallow collective bargaining over most non-wage benefits and workplace conditions” (Biggs and Richwine 2011, 4).¹

In short, Biggs and Richwine’s analysis is incorrect, and, even if it were not, it does nothing to inform the debate over SB 5. Rather than focusing on erroneous conclusions that are a diversion from the issues raised by SB 5, policymakers can make use of data directly relevant to the debate—data already published in the February EPI paper.

The EPI findings and Biggs and Richwine’s response

An often repeated argument for ending collective bargaining is to bring public employee costs in line with the private sector. But the research documented in Are Ohio Public Employees Over-compensated? shows that state and local government employees in Ohio are not overpaid. Comparisons controlling for education, experience, organizational size, gender, race, ethnicity, citizenship, and disability reveal that employees of state and local governments (referred to in this paper as public employees) earn lower wages than comparable private-sector employees. Average annual wages and salaries of full-time state and local public employees in Ohio are 5.9 percent lower than those of comparable private-sector employees. However, some full-time public employees work fewer hours on average, particularly college-educated employees. When annual hours worked are factored in, full-time state and local employees earn 3.3 percent less in wages and salaries than similar private-sector workers.

Looking at total compensation (wages and nonwage benefits), Ohio public employees annually earn 6.0 percent less on average than comparable private-sector employees and 3.5 percent lesson on an hourly basis. The analyses were adjusted for the increases in employer contributions to pension and retiree health insurance from 12 percent to 15 percent of total compensation, which reduced the hourly total compensation comparison to zero. In other words, public employees in Ohio earn no more and no less than similar private-sector employees in the state.

These comparisons account for important factors that affect earnings, the most important of which is level of education. Because occupations in the public sector require much higher levels of education, Ohio public-sector workers, on average, are more highly educated than private-sector workers; 49 percent of full-time public-sector workers in Ohio hold at least a bachelor’s degree, compared with 26 percent of full-time private-sector workers. Ohio state and local governments pay college-educated employees 24 percent less in annual total compensation, on average, than private employers.

The comparisons also reflect a big difference between the mix of wages and nonwage benefits in public and private sector compensation packages in Ohio. State and local government employees receive a higher portion of their compensation in the form of employer-provided benefits, and the composition of benefits is different from that provided in the private sector. For example, 29.7 percent of state and local government employee compensation expenses are devoted to nonwage benefits, compared with 18.9 percent to 22.8 percent of private-sector employee compensation costs. Public employers devote a larger share of their compensation packages to health insurance and pension benefits. Health insurance accounts for 12.9 percent of state and local government compensation but only 7 percent to 9.5 percent of private-sector compensation. Retirement benefits, including retiree health benefits, also account for a substantially greater share of Ohio public employee compensation—15 percent compared with 2.5 percent to 4.9 percent in the private sector.

Most public-sector employees also continue to participate in defined-benefit pension plans managed by the state, while most private-sector employers have switched to defined-contribution plans, particularly 401(k) plans, and many private-sector employers do not provide any contribution toward their employees’ retirement compensation beyond Social Security. On the other hand, public employees receive considerably less supplemental pay and somewhat less vacation time, and public employers contribute significantly less to legally mandated benefits financed through payroll taxes. For example, Ohio state government employees do not participate in Social Security, nor do most municipal workers.

To summarize, while some benefits are more generous in the public sector, it would be wrong to conclude that comparability of compensation between the public and private sectors requires that every element of compensation be the same. What is important is this: Considering both the cost of employer-provided benefits and direct wages, public-sector workers in Ohio receive compensation that is equal to what they would receive in the private sector.

The analysis that produced the findings above was criticized by Biggs and Richwine (2011) on three counts: that it omits the costs of retiree health benefits, doesn’t account for the guaranteed nature of public-sector pensions, and doesn’t factor in the value of job security. According to Biggs and Richwine, when these alleged omissions are included, Ohio public employees are overpaid by 43 percent.

Valuing retiree health benefits

Most states have pay-as-you-go financing of retiree health care benefits. This means that each year a state must allocate funds from its operating revenue to pay for health care for retired public employees. Ohio does not. Instead, Ohio prefunds its retiree health benefits from employer and employee pension contributions, which are then invested, with the investment income substantially contributing to the provision of these benefits. Ohio is considered a model plan for other states that still rely on pay-as-you-go financing. The Government Accountability Office estimated that retiree health benefits cost states approximately 2 percent of salary or 1.5 percent of total compensation (U.S. GAO 2007). While many states are debating whether they should begin prefunding these benefits, Ohio already does it.

Biggs and Richwine (2011) criticized the EPI study analyzing Ohio public employee pay for failing to account for retiree health benefits. They are mistaken. Because the analysis includes employer contributions to pensions, and because these contributions are invested to help cover retiree health care, the EPI analysis does include the retiree health benefit contribution.

Furthermore, a basic premise of the accounting in the Biggs and Richwine study (2011) is false. Biggs and Richwine maintain that retiree health insurance is an irrevocable and unalterable right, which is mandated to be funded by the state irrespective of any changes in the labor force or the state’s finances. But in Ohio public-employee retiree health care is not a guaranteed benefit. As the Ohio Public Employees Retirement System website explains, the OPERS Board of Trustees has made a commitment to provide health insurance as long as funds are available. OPERS has a separate retiree health insurance fund, and OPERS reports that “health care funding progress is measured by solvency period. OPERS has accumulated funds for a solvency expectation of about 11 years for this discretionary benefit”(OPERS 2011a). The 2010 OPERS Post-employment Health Care fund reserve was $12,319,743,979 as of Dec. 31 (OPERS 2011a, 34).

Finally, Biggs and Richwine’s accounting of how many and what proportion of state and local employees and private-sector employees participate in the retiree health insurance is too high. The most reliable source of data on current practices is the Medical Expenditure Panel conducted by the U.S. Department of Health and Human Services. According to panel data (U.S. DHHS 2011) only 36.4 percent of state and local governments provide health insurance to retirees under age 65, with 25.4 percent providing health insurance to retirees over age 65.

While it’s true that most state governments (all of which are large employers) offer retiree health benefits, many large private-sector firms (with more than 1,000 employees) also continue to provide health benefits to retirees: 34.5 percent of private firms with more than 1,000 workers provide such benefits to those under age 65 and 31.8 percent of such firms provide them to those over age 65. Other reports estimate employee use of retiree health insurance by looking at current employees who may become eligible if they eventually meet the age and service requirements.

For example, Richwine and Biggs (2011) cite a Pew Center on the States study (2010) that reports that 82 percent of state and local government employees in governmental units larger than 200 were eligible to receive retiree health care. However, to be eligible does not mean that employees will eventually receive these benefits. Employees must qualify for benefits through minimum years of service. Nonetheless, given the high rate of health cost inflation in the United States, the future prospects for retiree health insurance are grim. Ohio, however, retains considerable discretion over these benefits, including the rights to change co-insurance payments and even to cancel these benefits as is being done in the private sector. Presently, a number of states are integrating their retiree health benefits plans with Medicare Part D, which should offer significant savings for retiree prescription drug plans.

In summary, since retiree health benefits are discretionary, separately pre-funded, and not as widespread as estimated, the criticisms presented by Biggs and Richwine are incorrect. As a result their assertion that the provision of retiree health insurance represents a 6.3 percent additional cost to employee compensation is wrong.

It should be noted that the Pew Center on the States, which has been highly critical of the states’ management of public employee pension management and retiree health plan funding, gave Ohio its highest rating for managing long-term liabilities for both pensions and retiree health care in its 2010 report. They praised Ohio as a national leader in managing its long-term liabilities, and classified Ohio as a “solid performer” with respect to its pension fund and post retirement benefits—one of only 16 states to achieve that distinction.

Valuing employer contributions to public employee defined-benefit pensions and Social Security

There is substantial disagreement in the pension literature about measuring the present value of future pension payouts. This disagreement revolves around the “discount rate”—the appropriate interest rate to use when calculating the amount of money needed to be set aside today in order to make promised payments to retirees in the future. This rate is important in the discussion of employee compensation because it impacts how the cost of the pension component of compensation is calculated. Biggs and Richwine use an unnecessarily low rate which inflates the amount of money needed to be reserved to cover pension obligations. They compound this flaw by overestimating what those pension obligations will amount to in the first place (by, among other things, double counting retiree health costs, as explained above).

Many financial economists prefer the “riskless rate” measured as the yield on a 10-year (2.01 percent) or 30-year (3.26 percent) U.S. Treasury Bond to discount future pension obligations.² In contrast, standard actuarial analyses based on generally accepted state and local accounting rules calculate liabilities using the historical return on plans’ return on assets, i.e., a discount rate of 8 percent, although this may be changing with the new Governmental Accounting Standards Board (GASB) guidelines.

In keeping with this standard, OPERS’ actuaries assume that over a 30-year period, the full working career of a public employee, the plan will earn 8 percent annually, on average. Over the past 30-year period, OPERS has exceeded that objective by earning on average 8.99 percent annually. Historically, investment returns fund approximately two-thirds of the income used to pay members’ Ohio public pension benefits. (Public employee defined-benefit pension plans in Ohio are funded by sizeable employer contributions of 14 percent to 18 percent of payroll (which includes 6 percent for retiree health insurance) and 10 percent to 11.1 percent of employee contributions.³

Despite OPERS’ 8.99 percent historical annual rate of return on long-term investments, Biggs and Richwine (2011) argue that because the Ohio public pension plans guarantee payments to retirees, the actuaries should be using a 4 percent “risk-free” interest rate to discount the pension plan’s long-term obligations. Citing an actuarial study commissioned by the state of Florida, they estimate that halving the projected rate of return (from OPERS’ assumed 8 percent to 4 percent) multiplies pension costs by an “adjustment factor” of 3.59, despite the fact that the Florida study does not actually estimate pension costs with an 8 percent discount rate.⁴ They then apply the adjustment factor to an erroneous pension normal cost of 15.44 percent of wages that double-counts retiree health costs, before subtracting the employee contribution of 10 percent. This procedure results in a pension cost equal to a whopping 45.4 percent of salaries.

In contrast to Biggs and Richwine, the OPERS trustees report a much lower employer normal cost of 5.35 of wages for the pension Traditional Plan (OPERS 2011a, 146). Even if we accept Biggs’ and Richwine’s dubious adjustment factor of 3.59, multiplying this by the trustees’ normal costs for the employer yields a pension cost of 19.21 percent, not 45.4 percent, of salaries.

This estimate is still predicated on an artificially low discount rate. Biggs and Richwine’s use of a 4 percent discount rate is based on the assumption that benefits are guaranteed and funding methods are fixed. In other words, this is the estimated long-run rate of return if OPERS wanted to guarantee that the amount contributed today would be sufficient to pay estimated future benefits and were willing invest in virtually risk-free assets such as U.S. Treasury bonds to ensure a fixed rate of return.

In fact, OPERS, like nearly all pension funds, invests in a balanced portfolio that includes stocks and corporate bonds among other assets. The risk-free rate is much lower than the expected rate of return on these assets, which OPERS estimates at 8 percent. If OPERS based required contributions on the risk-free rate rather than the expected rate of return on these assets, this would result in systematic overfunding and would shift pension costs from future taxpayers (who are likely to be wealthier due to economic growth) to current taxpayers.

Biggs and Richwine do not claim that the risk-free rate is the expected rate of return on pension fund assets, and Biggs has elsewhere assumed a higher rate of return for individual retirement accounts (Biggs 2002; Morrissey 2011). Instead, their argument is that a risk-free rate should be used because pension benefits are guaranteed. This is an incorrect assumption. While benefits accrued to date are guaranteed, a pension’s normal cost is based on projections that assume wage growth and other factors. So, for example, if pension benefits are based on years of service and an employee’s final average salary, pension funds estimate the normal cost based on a projected future salary rather than the employee’s current salary, though the employee is guaranteed neither continued employment nor salary increases (and employers facing unexpectedly high pension costs are more likely to constrain pay growth and lay off workers).

Furthermore, the trustees can adjust not only employer but also employee contributions to meet obligations if the realized rate of return is different than the projected rate of return. In the last several years, for example, employee contributions to OPERS increased from 8.5 percent of wages to 10 percent of wages in response to the financial collapse of Wall Street in 2008, and OPERS has done well to recover most of those investment loses (OPERS 2011a). Selecting the appropriate discount rate is critical in pricing any long-term project or investment. Since investment returns cover roughly two-thirds of the cost of OPERS pension benefits, using the appropriate discount rate is essential in evaluating the plans’ current and future performance. Table 1 illustrates the impact of alternative discount rates.

Table 1

As the table shows, the selection of the appropriate discount rate greatly influences the valuation of $10,000 invested annually over 30 years. What the table does not show is the likely increase in volatility in returns as the discount rate rises. The advantage of a defined-benefit pension plan is that it creates a perpetual investment fund that can diversify against volatility over a long period of time (for example, an employee’s career of 30 or 40 years) in order to invest in securities that are more likely to produce higher returns.

In the past 30 years OPERS exceeded its investment goal by 20 percent, earning 2.43 times more than they would have earned with a 4 percent risk-free investment. In the past decade, OPERS, like most pension plans, has been unable to achieve its investment goals due to two serious recessions and a global financial crisis. Nonetheless, while earning only a 5 percent annual return during this last difficult decade for investors, OPERS has managed its portfolio well and should be able to achieve its goals over the next 30 years. A discount rate of 4 percent as applied by Biggs and Richwine clearly understates the value of Ohio’s pension assets.

A quick note about Social Security

As with pensions and retiree health insurance, Biggs and Richwine (2011) lose their focus on the cost of employer contributions to Social Security. They correctly report that private employers contribute 6.2 percent of payroll for Social Security taxes for wages up to $106,800 paid in 2011. But they then arbitrarily value that contribution at 2 percent.⁵

This is illogical. Comparing private and public employee earnings requires comparing employer costs for total hourly compensation, controlling for a variety of human capital attributes (education, experience, etc.). Compensation packages of equal cost are equal regardless of how compensation is allocated across wages and benefits. Thus the task is to compare employer costs of compensation, not what employees do or what others do on behalf of the employees with their compensation. The employer contributions to Social Security are part of the cost of the employees’ compensation. The EPI study accounts for the full Social Security contributions private-sector employers make on behalf of their employees in its benefits markup, based on the Employer Costs for Employee Compensation (ECEC) survey data collected by the Bureau of Labor Statistics (U.S. BLS 2010). Richwine and Biggs (2011) use their misrepresentation in accounting for employer contributions to Social Security as an arbitrary device to reduce the private-sector employers’ compensation costs by 4 percent, which inflates their estimated relative value of public employee pay.

Misapplying a compensating earnings differential for earnings instability

Biggs and Richwine (2011) assert that Ohio public employees receive an unwarranted compensation premium arising from job security. It is important to make a distinction between job security, which involves a legal guarantee of a job, and job stability, which arises from the normal working of the labor market. Ohio public employees have had job stability, but they do not have job security. More than 40,000 Ohio public employee jobs have been lost in the last five years (U.S. BLS 2011a), which clearly demonstrates that Ohio public employees lack any legal guarantee of job security.

Failing to make the distinction between security and stability leads Biggs and Richwine (2011) to misapply the finance concept of a “certainty equivalent” in building their model of job security to estimate a compensating earnings premium of 1.8 percent. Biggs and Richwine (2011) say that the utility model that they built shows that this increased public employee job security is equal to a 1.8 percent increase in compensation, based on revealed preference theory, i.e., that employees act “as if” they value a preference without actually demonstrating they do value an outcome or preference. This is a model based on a theory of job security that has no empirical support. In their discussion of job stability among Ohio private and public employees, they also conceptually shift from involuntary job loss to unemployment as their measure of instability, but unemployment is primarily composed of entrants to the labor force and is therefore not a accurate measure of job instability.

Instead of building a hypothetical model, we will directly examine the empirical evidence on job stability and compensating wage differentials. The Bureau of Labor Statistics’ Job Openings and Labor Turnover Survey (JOLTS) data (U.S. BLS 2011b) indicate that in 2010 the average private-sector worker had an 18 percent chance of being discharged or laid off during the year while the average state and local public employees had a 7 percent chance of involuntary job loss. The private sector has a wide range of practices; for example construction workers had a 57 percent change of involuntary job loss, while workers in finance and insurance had the same 7 percent rate as state and local public employees in 2010.

Empirically, if job stability is as valuable as Biggs and Richwine claim, we should be able to observe its effects on earnings across industries. Using the JOLTS data for 14 private-sector industries, average annual rate of discharge and layoff is averaged over the years 2001 to 2009 and reported in Table 2 in column one. We use the largest employing sector, retail, to make our relative comparisons. With a 20 percent annual rate of discharges and layoffs, retail has a rate slightly higher than the average for the entire private sector. If employment stability is as highly valued as Biggs and Richwine claim, there should be an observable compensating wage differential, all else equal, reflected in a large and observable pay premium, to attract workers to high involuntary turnover industries, while those who work in lower than average involuntary turnover industries should receive a large pay penalty, similar to what Biggs and Richwine predict for government workers.

Table 2

Specific predictions are based on the rankings and rates of involuntary turnover by industry: those experiencing an involuntary turnover rate greater than 20 percent should have an instability premium, while those with a turnover rate less than 20 percent should have a stability penalty. If there is a meaningful compensating job stability differential, workers in the arts, entertainment, and recreation industry, for example, should receive the largest instability premium, since 49 percent of them on average experience an involuntary turnover per year, whereas workers in finance and insurance should receive the largest earnings penalty, since their annual involuntary turnover rate is only 9 percent.

A private-sector earnings equation with data from the BLS’ Current Population Survey with a complete set of human capital controls was estimated, so as not to confound the results with government earnings. The industry-omitted variable is retail, so the results reported in column two are in relation to the retail sector. Column three reports the variance of the estimated actual earnings differential from the predicted earnings differential. As is apparent in column three, without the need for any statistical tools, the predictions are uncorrelated with the results.

There is no obvious industry-level compensating wage differential for employment instability. The finance and insurance industry, which should have the largest job-stability penalty, has the third largest premium,⁶ whereas the accommodation and food services industry, which should have a large instability premium, in fact, has the largest penalty. It is not surprising that the estimates for a compensating wage differential premium/penalty failed.

Compensating differentials are notoriously difficult to capture. While most labor economists believe that compensating wage differentials do exist, there are many offsetting factors. Jobs are bundles of tasks and attributes that make them simultaneously both attractive and unattractive. Stability may be offset by a lack of control, autonomy, authority, safety, desirable work environment, commuting distance, or flexibility. There is no observable job instability compensating wage differential.

In the case of Ohio, our empirical results indicate that public employees are compensated roughly equal to private-sector employees. Biggs and Richwine, however, believe that Ohio public employees receive compensation approximately 31.2 percent higher than they would be likely to receive in alternative private-sector employment, because of their upward adjustments to hypothetical—not actual—employer contributions to pensions and retiree health benefits. Since they believe that job security protects not just the public employee job, but the pay premium as well, they incorporate their calculated 31.2 percent premium into their utility model and their estimated value of job security increases from 1.8 percent to 9.3 percent of compensation. Because we find no evidence of a compensation premium from job stability, we reject Biggs’ and Richwine’s assertion that public employees receive a 9.3 percent stability premium, and we reject their contention that Ohio public employees are overpaid by 43 percent.

The nature of the labor market for state and local government employees offers more plausible explanation for job stability. Approximately 49 percent of Ohio public employees are college educated. Americans with a college education had an average annual unemployment rate of 3.1 percent (ranging from 2.0 percent to 4.7 percent) from 2001 to 2011, while those with a high school education had an average unemployment rate of 6.3 percent (ranging from 4.2 percent to 10.7 percent) for the same period (U.S. BLS 2011c). A low unemployment rate labor market, indicating labor scarcity, requires organizations to devise different human resource policies regarding retention and turnover than a labor market that provides an ample supply of qualified labor.

Furthermore, state and local government employees actually experience much higher rates of layoffs than are recorded in the data. Most full-time employees working in education (kindergarten through university), who account for 54 percent of state- and local-government employment, are on 10-month contracts that provide extended 10-month salary payments and health benefits to be paid over 12 months. These teachers, faculty, and support staff represent more than 4 million employees nationally whose employment is interrupted each year but who are not subject to formal layoffs, which provides each state enormous savings in payouts for unemployment benefits that other seasonal workers often depend upon. As a consequence of the structure of their 10-month employment contracts, these public workers do not appear in the layoff data each year. Public employees experience instability in employment that is unaccounted for in the data and therefore understated in any employment stability analysis.

Conclusion: Public employees are not overcompensated

The earnings equation estimates from the EPI paper Are Ohio Public Employees Over-compensated? indicate that state and local government employees are not overpaid. These findings remain sound. Comparisons controlling for education, experience, hours of work, gender, race, ethnicity, citizenship, and disability find that public-sector wages and benefits are less than private-sector wages and benefits. There is no significant net difference between private and public employee compensation costs.

The analysis presented in this new EPI paper demonstrates that the criticisms that the earlier research failed to account for retiree health benefits, guaranteed pension benefits, and employment security is erroneous. The prior research has accounted for these issues, and represents a much more accurate assessment of public employee compensation costs, as summarized below.

EPI’s analysis accounts for retiree health benefits; Biggs and Richwine double counts them

Public employee retiree health benefits in Ohio are prefunded through employer pension contributions, and are discretionary benefits, not guaranteed by the state. The cost of retiree health insurance was included in the ECEC pension benefit contributions in the earlier analysis. The assertion that retiree health benefits represent an additional unaccounted for 6.3 percent cost to the state of Ohio is erroneous.

EPI’s analysis calculates employee pension costs based on a realistic rate of return; Biggs and Richwine inflate the cost by using a “risk free” rate of return

Public employee defined-benefit pension plans in Ohio are funded by sizeable employer contributions of 14 percent to 18 percent of payroll (which includes 6 percent for retiree health insurance) and 10 percent to 11.1 percent of employee contributions. Over the last 30 years OPERS has earned an 8.99 percent rate of return, exceeding its target rate of 8 percent by a 20 percent accrued value. Biggs and Richwine’s assertion that the plan should use a 4 percent “risk-free” interest rate to discount the pension plan’s long-term obligations because Ohio public pension plans guarantee payments to retirees and have inflexible funding is incorrect. While benefits accrued to date are guaranteed, a pension’s normal cost is based on projections that assume wage growth and other factors arising from wage and employment years that are not guaranteed. Furthermore, OPERS has the authority to increase contributions and use the retiree health-benefit plan contributions to meet pension obligations. By halving the projected rate of return and multiplying an erroneous normal pension cost that double counts retiree health costs by an unsubstantiated “adjustment factor,” Biggs and Richwine conclude that public employee pension costs equals 45.4 percent of salaries. This hypothetical cost is inflated and does not reflect current employer costs of employing an Ohio state or local employee.

Biggs and Richwine arbitrarily deflate private-sector Social Security costs

Biggs and Richwine’s treatment of the cost of Social Security payments is a manipulation to reduce private-sector compensation and widen the alleged compensation gap. They disagree with the funding mechanism for Social Security that has been in place since 1935. They arbitrarily decided that the employer’s 6.2 percent contribution is worth only 2 percent to employees. Since Ohio does not pay Social Security taxes for public employees, this lowers private-sector compensation costs by 4.2 percent and widens the alleged pay gap between private- and public-sector workers. This 4.2% adjustment to private-sector compensation costs has no logic in what is being computed—the cost of employing an employee. Private-sector employers who pay Social Security taxes do not receive the Biggs and Richwine adjustment.

There is no evidence for Biggs and Richwine’s job stability premium for public-sector workers

Although our comparison of job stability and pay in various industries shows no evidence of a job stability premium, Biggs and Richwine say that public employees earn a 9.3 percent compensation premium. This calculation is based on the faulty hypothetical upward adjustments to employer contributions to pensions and retiree health benefits discussed above, that produced an alleged 31.2 percent compensation premium. But they offer no evidence for a job stability premium for public employees in Ohio or elsewhere.

None of the issues raised by Biggs and Richwine are subjects of public employment collective bargaining in Ohio, and thus should not divert us from what is important in the public debate over Ohio Senate Bill 5.

Our research informs that debate. It revealed substantially different approaches to staffing and compensation between the private and public sectors. On average, Ohio state and local public-sector workers are more highly educated than the private-sector workforce; 49 percent of full-time state and local public-sector workers have at least a four-year college degree, compared with 26 percent of full-time private-sector workers. For workers with a bachelor’s degree, state and local governments pay salaries on average over 25 percent lower than private employers. The total compensation of college-educated private-sector employees is more than 20 percent higher than the compensation of similarly educated public employees. The earnings differential is greatest for professional employees such as lawyers and doctors. These earnings differences may create opportunities for cost saving by reviewing professional outsourcing contracts to examine what work might be performed by lower cost public employees.

The public sector appears to set a floor on compensation that raises the compensation of workers with high school educations in comparison with similarly educated workers in the private sector. This result is in part because the compensation floor has collapsed in the private sector.

Benefits are allocated differently between Ohio private- and public-sector full-time workers. State and local government employees receive a higher portion of their compensation in the form of employer-provided benefits, and the mix of benefits is different than in the private sector. Public employers underwrite 34.1 percent of employee compensation in benefits, whereas small private employers devote 26.3 percent of their compensation for benefits for employees, versus 33.1 percent for large employers. Ohio public employers provide better health insurance and pension benefits. Health insurance accounts for 7.4 percent of private-sector compensation but 11.2 percent of state and local government employee costs, a 50 percent greater share of employer costs. Retirement benefits also account for a substantially greater share of Ohio public employee compensation, 15 percent compared with 9.2 percent in the private sector including Social Security. Public employees also continue to participate in defined-benefit plans managed by the state, while private-sector employers have switched to defined-contribution plans, particularly 401(k) plans. On the other hand, public employees receive considerably less supplemental pay and less vacation time, and public employers contribute significantly less to legally mandated benefits.

A comprehensive assessment of the total costs of employing Ohio state and local government employees reveals that they are neither overpaid nor underpaid when compared to comparable private-sector employees.

—Jeffrey H. Keefe is associate professor of labor and employment relations at the School of Management and Labor Relations, Rutgers University, where he conducts research on labor markets, human resources, and labor-management relations to inform public policy. He teaches courses on collective bargaining, negotiations, financial analysis, benefits and social insurance, and strategic research. He earned his Ph.D. at Cornell University.

Endnotes

1. It also eliminates most dispute resolution procedures and removes the right to collective bargaining from a substantial portion of the public-sector workforce.
2. The current risk free rate is set by the yield on a 10-year Treasury bond, which is currently 2.01, as of September 12, 2011. This risk free rate should be used by Biggs and Richwine; however, the use the conventional risk free rate of 4%, which is completely arbitrary. If the used the 2.01 percent rate, the unfunded liabilities would be so large that no one could accept their estimates.
3. Consulting firm Aon Hewitt (2011) reports that large employers provide 13.8 percent of compensation toward retirement which includes Social Security, while the state of Ohio contributes 14 percent.
4. The letter from actuarial and consulting firm Milliman (DuZebe 2011) that they use to calculate the adjustment factor does not provide an analysis of an 8 percent target rate and thus does not indicate a 3.59 adjustment factor. It provides an analysis of 7.75 percent target rate for Florida. Using the 4 percent rate, Milliman provides an adjustment factor of 2.97. Biggs and Richwine do not provide documentation on how they arrive at a 3.59 adjustment factor.
5. Biggs and Richwine use the 2 percent figure because they disagree with Social Security as a pay-as-you-go plan, even though it provides the majority of income for more than 80 percent of the retired workforce—an arbitrary and incorrect decision.
6. In 2010 and 2011, the finance and insurance industry had the lowest involuntary turnover in the United States, considerably lower than public employment, while providing the highest compensation after controlling for human capital, strongly suggesting there is not a powerful compensating differential for stability operating in U.S. labor markets.

References

Aon Hewitt. 2011. “Aon Hewitt Benefit Index: Summary of 2010 results for Ohio PERS.” PowerPoint presentation accessed from Ohio Public Employees Retirement System website. https://www.opers.org/news/ORSC/2011/Aon-Hewitt_Benefit_Summary.pdf

Biggs, Andrew G. 2002. “The Stock Market and Social Security Reform.” Cato Institute website, Sept. 21. http://www.cato.org/pub_display.php?pub_id=3601

Biggs, Andrew and Jason Richwine. 2011. Public vs. Private Sector Compensation in Ohio: Public workers make 43 percent more in total compensation than their private‐sector colleagues. Columbus, Ohio: Ohio Business Roundtable.

DuZebe, Robert S. 2011. Study Reflecting Impact to the FRS of Changing the Investment Return. Seattle: Milliman.

Keefe, Jeffrey H. 2011. Are Ohio Public Employees Over-Compensated? Economic Policy Institute Briefing Paper # 296. Washington, D.C.: EPI.

Morrissey, Monique. 2011. Discounting public pensions: Reports of trillions in shortfalls ignore expected returns on assets. Economic Policy Institute Policy Memorandum #179. Washington, D.C.: EPI. http://www.epi.org/page/-/EPI_PolicyMemorandum_179.pdf

Ohio Public Employees Retirement System (OPERS). 2011a. The Comprehensive Annual Financial Report 2010. Columbus, Ohio: OPERS.

Ohio Public Employees Retirement System (OPERS). 2011b. 2010 Summary Annual Financial Report. Columbus, Ohio: OPERS.

Pew Center on the States. 2010. The Trillion Dollar Gap: Underfunded State Retirement Systems and the Road to Reform. Washington, D.C.: Pew Center on the States.

Richwine, Jason and Andrew Biggs. 2011. Are California Public Employees Overpaid? Heritage Foundation Center for Data Analysis Report No. CDA11-01. Washington, D.C.: Heritage Foundation; http://www.heritage.org/ Research/Reports/2011/03/Are-California-Public-Employees-Overpaid.

U.S. Bureau of Labor Statistics (BLS). 2010. “Employer Costs for Employee Compensation, June 2010,” BLS website (accessed March 2011.) (With unpublished detailed compensation data for the East North Central Census Division).

U.S. Bureau of Labor Statistics (BLS). 2011a. “Current Employment Statistics. State and Area Employment, Hours, and Earnings.” BLS website accessed Sept. 20.

U.S. Bureau of Labor Statistics (BLS). 2011b. “Job Openings and Labor Turnover Survey” (JOLTS). BLS Office of Employment and Unemployment Statistics website, accessed Sept. 20. http://www.bls.gov/jlt/

U.S. Bureau of Labor Statistics (BLS). 2011c. “Current Population Survey, annual averages, household data,” BLS website accessed Sept. 20. http://www.bls.gov/cps/tables.htm#charunem_m

U.S. Government Accountability Office (GAO). 2007. State and Local Government Retiree Benefits: Current Status of Benefit Structures, Protections, and Fiscal Outlook for Funding Future Costs. Report to the U.S. Senate Committee on Finance. Washington, D.C.: GAO; http://www.gao.gov/new.items/d071156.pdf

U.S. Department of Health and Human Services (DHHS)/Agency for Health Care Research and Quality (AHRQ) and National Center for Health Statistics (NCHS) Medical Expenditure Panel Survey (MEPS). Accessed March 10 2011.

This report estimates the economic and job creation impact of a major investment in water infrastructure in the United States. This number—$188.4 billion—is based on the level of investment necessary, as estimated by the Environmental Protection Agency, to manage stormwater and preserve water quality across the country. An investment of $188.4 billion spread equally over the next five years would generate $265.6 billion in economic activity and create close to 1.9 million jobs.