Mitchell’s Musings 12-19-16: Who Will Play George Shultz?
Daniel J.B. Mitchell
Last week’s musing dealt with economic data and possible dangers to the integrity of the federal data collection and distribution system in the event of a rogue president. In that musing we noted the episode in which former President Richard Nixon believed that a conspiracy among federal data analysts was presenting economic news unfavorable to him.
Although we tend to think of newsworthy economic data as a fairly narrow group of series - inflation numbers from the CPI and other indexes, the unemployment rate, and real GDP growth – the federal data agencies turn out a very large array of numbers regularly. Even if these figures don’t make the news, they do influence the perceptions of forecasters and others who then interpret them in ways which filter into the larger public understanding of how well or poorly the economy is progressing.
There are various aspects of economic data that are involved. Many series – such as the Big 3 mentioned earlier (inflation, unemployment, real growth) – are represented by specific series that are linked to concepts that have to be defined for measurement. For example, while inflation is generally defined as broad-based price increases, in practice there are detailed methodologies of data collection and data combination that are meant to reflect that concept as a specific index. Changes in methodology can produce variation in the series. There are in fact several versions of the CPI, to take an example, which differ in the “headline” number they produce due to variations in weighting and other factors.
Another important element is the timing of data releases. It is the practice of the major statistical agencies to release data on an announced-in-advance schedule. This practice has been in use at least since the 1960s. The notion is that with pre-announced timing, there won’t be charges that particular numbers were either withheld for a time (perhaps bad news for an incumbent) or rushed out (good news) to influence elections or other political decisions.
In fact, during the Nixon administration – even apart from the President’s conspiracy views – there was controversy about how fast the numbers should be released. Should the numbers just appear when ready? Should statistical officials say something about them once released, i.e., provide some interpretation? Or should commentary be left to political appointees and outside analysts? That issue had also arisen in the Lyndon Johnson administration after an episode in which the President had commented on what-he-viewed-as favorable statistical news before the official release date.
There are often important policy decisions that are linked to economic data. The Federal Reserve, for example, looks at the general macro condition of the economy before making its monetary decisions. Social Security payments are indexed to the CPI. It is inevitable, therefore, that key data series will be in danger of politicization and ultimately their protection will depend on key high officials.
In the Nixon case, the protector role was played by George Shultz who was first Secretary of Labor and then Secretary of the Treasury. Back in a Mitchell’s Musing of October 22, 2012, I noted Shultz’s role as a stabilizer of Nixon when the latter wanted to do something that he shouldn’t. In the case described in that Musing, Nixon wanted to leak in advance certain Department of Commerce figures which he thought were favorable. Shultz, however, after a phone conversation with Nixon, effectively ensured that the numbers were released appropriately.
Shultz had a background as a professional economist and academic researcher. I suspect that most persons with that background would have the same inclinations as Shultz regarding data integrity. We have only limited information at this time as to who will be Trump’s key economic advisors. Gary Cohn, Trump’s designee as chair of the National Economic Council, is President and Chief Operating Officer of Goldman Sachs. Trump’s Secretary of the Treasury designee is Steve Mnuchin, also a person with a Goldman Sachs background (and other financial ventures). His Secretary of Labor designee is Andrew Puzder, a fast-food chief executive. (Within Labor is the Bureau of Labor Statistics.) And his Secretary of Commerce is Wilbur Ross, an investor who is said to be a turn-around specialist. (Within Commerce is the Bureau of Economic Analysis – producer of the national income accounts – and the Bureau of the Census.)
The only person with an academic/research background connected with Trump is Professor Peter Navarro of the University of California, Irvine. However, it is unclear what that connection is – Navarro seemed to have attached himself to Trump during the election campaign initially rather than having been first invited. In any case, Navarro has so far not been named to any position in the Trump administration. Lawrence Kudlow, a former Reagan appointee with a background in the financial world and CNBC, is reported to be in line for chair of the President’s Council of Economic Advisors (CEA). The CEA has traditionally been a location of academic researchers, but it will apparently not be this time, at least at the top.
In short, there does not seem to be anyone at this point who could play the role in the forthcoming Trump administration that Shultz did with Nixon. There could thus be trouble ahead, particularly if any official data series shows an unfavorable result.
 For example, the release data in 2017 for the U.S. Bureau of Labor Statistics are listed at https://www.bls.gov/schedule/2017/home.htm.
 Joseph P. Goldberg and William T. Moye, The First Hundred Years of the Bureau of Labor Statistics, 1985, pp. 221-226. Available at http://digitalcommons.ilr.cornell.edu/cgi/viewcontent.cgi?article=1699&context=key_workplace.
 Ibid, p. 188.
 It’s also important to note that Shultz in 1968 was president of LERA (the host site of EPRN and thus of this Musing); LERA was then known as the Industrial Relations Research Association.