Discredited research doesn’t discredit need for cuts

You can appreciate, respect and value politics — and still find it difficult to live with sometimes.

One of those “sometimes” occurred recently when an error was discovered in an academic paper published three years ago in the American Economic Review.

Very few Americans have ever read an article in the American Economic Review. It is not aimed at the general public but is one of a small handful of “must read” professional journals for economists.

The research papers it publishes vary widely in the subjects covered but often are of very narrow scope, of interest primarily to specialists. They also vary in their readability; some are fairly accessible to general readers, others look like an MIT sophomore’s nightmare math final.

The May, 2010 article that started the dustup, “Growth in A Time of Debt” by Harvard University professors, Catherine Reinhart and Kenneth S. Rogoff, was one of the more accessible articles. It was short — six pages — and the statistics on which it was based were displayed clearly enough for readers to get the point.

The overall conclusion was that the growth of public debt has a negative impact on economic growth. Unfortunately, the authors also chose to note on the first page that, “for levels of external debt in excess of 90 percent of GDP, growth rates are roughly cut in half.” This was exactly the kind of easily remembered magic number that politicians thrive on.

It was not wrong for the authors to point out this number, for it appeared to be the fiscal equivalent of the “Plimsoll Line,” the maximum safe loading point for ships, after which they become unstable. In the clarity of hindsight, though, it was a mistake to include the 90 percent reference on the first-page summary for it ended up diverting attention from the research and its principal conclusions.

It was not Reinhart and Rogoff’s only mistake. Two economics professors from the University of Massachusetts, with the aid of a researcher, discovered that the data underlying the debt-growth relationship contained an error. In transforming the source data into spread-sheet format, the first five observations were dropped. Oops.

There was, and still is, an academic debate over the error, and it should surprise no one that the debate includes arguments over whether the error was accidental or deliberate. Welcome to politics, 2013.

The temperature of the debate reflects more than the traditional stimulus vs. fiscal restraint arguments as modern echoes of the views of John Maynard Keynes vs. those of Friedrich Hayek. The pro-stimulus side has generally abandoned the economic argument for stimulus spending and seems focused instead on discrediting the idea of what they call austerity — government spending cutbacks. The error in this research paper gave them what they believed was the perfect sword to slay the austerity dragon.

That belief was mistaken, though. Reports of austerity’s death turned out to be exaggerated.

The reason is simple enough. Austerity, like stimulus spending comes in many different flavors. There is a big difference, for example, between an austerity program dictated by a country’s creditors and an independently developed government plan to reduce its dependence on unsustainable debt.

We know from historical private sector bankruptcy experience that a forced austerity program will not always address the problems that caused the financial crisis the first place. Similarly, a misguided austerity program may fail to right a country’s financial ship and, in some circumstances, prolong the economic misery of a downturn.

On the other side of fiscal policy, as we saw with our own Stimulus I in 2009, an ill-conceived stimulus program can be totally ineffective at prompting economic growth, employment or private investment. As happened with that program, stimulus spending may instead end up as a kind of mammoth employment security program for some workers. To the extent that it preserved jobs during the difficult early days of the recession, it may not have been totally useless. But it most certainly was not an effective economic stimulus.

From an economic policy standpoint, the real problem is less about economics than it is about allowing politics to define the economic terms used and the economic theory applied. Austerity, for example, has come to mean any limitation on the growth, or even the rate of growth, of government spending. This is not Keynesian economics; it is the worst sort of taxpayer-financed electioneering.

At bottom, the “Growth in a Time of Debt” was a research paper, nothing more or less. Much of its real merit lies in the questions it raises about the effects of spending or austerity on real economies. And ultimately the spending vs. austerity debate will not be resolved in economics until we uncover the answers to those questions.

James McCusker is a Bothell economist, educator and consultant. He also writes a monthly column for the Herald Business Journal.