Commentary | Jobs and Unemployment

David Neumark’s critique of EPI’s broadband report completely misses the mark

Download PDF

A new report from Sprint criticizes a recent EPI report on the jobs impact of wireless broadband infrastructure investment in the context of the AT&T/T-Mobile merger. The Sprint report, by University of California-Irvine economics professor David Neumark, claims that EPI’s findings are “completely unfounded,” but it is clear from Neumark’s critique that he likely has not actually read the EPI report. First, he got our name wrong, referring to EPI throughout as the “Employment Policies Institute.” We are, in fact, the Economic Policy Institute, and have been proudly so for the last 25 years. Our name is prominently displayed on our website and at the top of every report we publish—including the report he critiqued.

Second, he critiques findings that the paper never makes. He notes, “The EPI analysis claiming that the AT&T/T-Mobile merger will create jobs because of increased capital investment is completely unfounded.” The rest of his paper is dedicated to listing all the possible reasons why the merger might not result in additional net investment.
Yet even a casual glance at the report would reveal that it did not make such a claim. The report is titled “The Jobs Impact of Telecom Investment,” and its first sentence explains, “This memo describes the jobs impact of a prototypical investment in wireless infrastructure.” The report then finds that a $1 billion investment in wireless infrastructure would create approximately 12,000 job-years of work—a calculation that applies equally to AT&T, Sprint, T-Mobile, and any other telecom company that makes wireless infrastructure investments.

At the end of the report, we note that AT&T has claimed that it plans to increase net investment in wireless broadband infrastructure by $8 billion over seven years as a result of the proposed merger with T-Mobile. The report did not analyze the veracity of this claim, but found that these plans would create between 55,000 and 96,000 job-years if they were to come to fruition. At no point in his critique does Neumark dispute any of these findings.

The report did not examine whether or not the merger actually would result in a net increase of investment. Nor did it analyze the broader impact of the merger, which would involve extensive knowledge of what would happen under a scenario in which the merger did not occur. Could T-Mobile survive as an independent company, or would Sprint succeed in merging with T-Mobile (as it has already tried to do)? In each of these scenarios, what would be the investment levels? What would be the changes in the cost structure? This type of counterfactual analysis goes way beyond the scope of EPI’s report.
Our report addresses a key issue in the discussion of the merger—the impact of building out of our broadband infrastructure. Whether AT&T will make the investments is worth analyzing. My colleague Robert Scott has separately argued that it is plausible that AT&T will make the proposed infrastructure investments.i But policymakers can remove any uncertainty about these investments simply by legally requiring AT&T to make the investments as part of the federal government’s approval of the merger. This would ensure that AT&T’s promises become reality.
But we have released no report on whether—absent legal requirements—the net investment increase will in fact occur. Neumark’s attempted critique is thus a lesson to us all that it is always a good idea to read a report before you write a 20-page paper disagreeing with it.

i Robert Scott, “The jobs impact of the AT&T-T-Mobile merger, redux,” July 7, 2011 comment to Beth McConnell, “U.S. Workers Can’t Afford the AT&T – T-Mobile Merger,” June 27, 2011. Media and Democracy Coalition News blog,

See related work on Jobs and Unemployment

See more work by Ethan Pollack