In Monday’s Wall Street Journal, Andrew Biggs and Sylvester Schieber cited these statistics from the Organisation for Economic Co-operation and Development (OECD):

“Despite a supposedly stingy Social Security program and ineffective retirement-savings vehicles, the average U.S. retiree has an income equal to 92% of the average American income, handily outpacing the Scandinavian countries (81%), Germany (85%), Belgium (77%) and many others.”

Meanwhile, in its Global AgeWatch Index released Tuesday, HelpAge International ranked the United States #8 among the best countries to grow old in, ahead of France (#18) but trailing Norway, Sweden, Switzerland, Canada, Germany, Netherlands, and Iceland (#1-7). Afghanistan (#96) was in last place.

It’s not hard to imagine how wealthy countries like Norway and the United States outrank poor and war-torn countries like Afghanistan. But the relative ranking of the wealthy countries comes as a surprise. How did the United States and other English-speaking countries like the United Kingdom and Australia, not known for their generous social insurance programs or employee benefits, come close to the Nordic cradle-to-grave welfare states and handily beat out France, with its famously generous pensions and high-quality affordable healthcare? Are older Americans really living in a retiree paradise?

Hardly. A closer look at the measures used to rank the countries show that they are not a useful way to differentiate among wealthy countries in their treatment of seniors. The only question about pensions, for example, measures coverage rates—near-universal in most wealthy countries—not income replacement rates.

Ideally, the index should take into account the resources countries have at their disposal and consider how effectively and fairly those resources are used. To do so, measures should control for per capita income, or compare how seniors in each country fare relative to the general population. Only three of the 13 measures used in HelpAge’s Global AgeWatch Index take this approach: the relative income measure cited by Biggs and Schieber, a relative poverty measure, and a measure comparing the share of people over age 50 to the share of those aged 35-49 who said their life had an important purpose or meaning.

The poverty measure is also in the OECD’s Pensions at a Glance report, but Biggs and Schieber don’t mention it, presumably because the United States ranks very low among OECD countries in this key measure. Though Americans aged 66-75 have a relative poverty rate close to that of the general population thanks to a combination of earned income and Social Security, those older than 75 have a very high relative poverty rate (24 percent) compared to the general population (17 percent) and to seniors older than 75 in other OECD countries (14 percent).

One measure used in the HelpAge index is simply per capita GDP, which is just stacking the deck. Three more are tied to per capita GDP: life expectancy at age 60, healthy life expectancy at age 60, and the share of the population aged 60 and older with at least a secondary education. The United States ranks high in per capita GDP and educational attainment but low among wealthy countries in life expectancy and health measures.

The French score is weighed down by the fact that only 45 percent of French people aged 55-64 are employed compared to 61% of Americans the same age. This reflects both a higher overall unemployment rate due to Eurozone austerity measures and a lower retirement age.* Though it is frequently argued that French pensions are overly generous and contribute to a lower per capita GDP—a measure already factored into the rankings—it’s hard to make a case that generous pensions are bad for retirees. Moreover, HelpAge uses per capita GDP as “a proxy for the standard of living of people in a country,” but this is not an objective measure of well-being because it doesn’t factor in the number of hours worked (and therefore the loss of leisure) which is higher in the United States. A better measure is productivity per hour, which is only slightly higher in the United States than in France.**

Many of the remaining metrics are subjective measures of wellbeing, which are influenced by social norms about expressing dissatisfaction. The French, in particular, are famous for grumbling on surveys. Thus, when fewer French people over age 50 say they’re satisfied with their local public transportation system or feel safe walking alone at night—two measures used in the HelpAge ranking—this doesn’t mean the United States has a better public transportation system or lower crime than France (on the contrary, Americans are much less likely to use public transportation and much more likely to be victims of violent crime).

This brings us back to the statistic cited by Biggs and Schieber, which turns out to be directly related to additional hours worked—not by prime-age workers this time, but by seniors. Though Biggs and Schieber refer to “retirees,” the measure they cite is the relative income of people 65 and older—retired or not—to that of the overall population. The reason the relative income of seniors is higher in the United States than in most OECD countries is simply that more U.S. seniors are still working, both because many are still relatively young (a greater share are in their 60s) and because the labor force participation of older people at any given age tends to be higher in the United States than in most advanced economies.

The upshot? These measures tell us that the United States is a wealthy country, in part because Americans work more hours and longer into old age than people in other countries. They also tell us that Americans tend to be upbeat in responding to surveys. They do not tell us much about how U.S. retirees compare to workers—excluding “retirees” who are still working—or how income is distributed within these groups. The fact remains that Americans are more likely to experience significant declines in their standard of living when they retire and to be relatively poor in old age than people in other advanced economies, and this is only expected to get worse.

*The early eligibility and normal retirement ages in France are gradually converging with those in the United States, but are currently lower in France for workers with long careers.

**The difference may not be statistically significant given the difficulty of accurately measuring work hours.