Seeing your research become part of conventional wisdom even outside the world of economics is certainly one of the highest honors an economist can hope for. Reinhard and Rogoffs research on the effects of financial crises is among those works. For a long time, their results that (systemic) financial crises tend to be followed by slow recoveries seemed like a little disputed fact inside and outside of academia. With the upcoming presidential election in the United States drawing ever closer, this appears to be changing. Yet a lot of the talk surrounding the issue reeks more of ideology than of cool-headed discussion.
The starting point of the whole debate appears to have been a new paper published by Bordo and Haubrich. They take another look at the performance of the United States economy following financial recessions, and their message is clear: we strongly disagree with R&R. This type of research is invaluable, for the danger of getting stuck within a framework and taking conclusions arrived at in the past as facts is all-too present in economics. Yet research needs to be judged by its merits and methods rather than by the convenience of the results. A brand new Ford Mustang, coincidentally painted in red and straight off the assembly line in Michigan, might seem appealing, but it is worth looking under the hood before buying it.
First, a quick rundown of the issue. Disruptions to the financial system tend to have a big impact on the real economy. The financial system, of course, is the link connecting the act of saving with the act of producing (or investing). It does not take a rocket scientist (or even an economist) to realize that a breakdown of this link leads to problems. Yet, as R&R noted, economies appear to remain stuck in recession long after the shock that caused the crisis has passed – recoveries after financial crises appear to be L-shaped. Possible reasons for this are many: wage and price stickiness, the zero lower bound to interest rates, debt overhang, income uncertainty and government intervention (which can go both ways), for instance. The American economy appears to be recovering slightly better than would have been expected in terms of historic comparison.
In come Bordo and Haubrich. And they claim exactly the opposite: the US recovery after the financial crisis in 2008 is dismal compared to other crises. So how do they arrive at such different conclusions? First, they ask a different question. R&R measure the strength of recoveries in terms of how long it takes for a country to return to its pre-crisis level of GDP/capita. B&H measure it as the “steepness” of the recovery after reaching the bottom of the crisis. Generally, the deeper the recession in the first place, the stronger the recovery. The argument they make for using this method is shaky at best and seems to lack consistency. It seems to make no sense to separate the depth of a recession from the way an economy behaves in the aftermath of it. They further use a different definition of financial crisis – one that makes them add, e.g., the FED-induced recession of 1981 to the list. Their analysis is also strictly confined to the U.S. economy, leaving valuable (yet admittedly hard to compare) data on the rest of the world out. Still, any analysis of the effects of financial crises that leaves Japan out of the picture is clearly missing something.
The political implications of the different results differ just as strongly as the results of the studies themselves. While the conventional take of the issue suggests that the current administration’s efforts during the crisis have held up fairly well, the logical conclusion to draw if taking B&Hs study at face value is that the president has been a huge failure. Yet the core of the issue is not the question about the record of an incumbent president, but whether everything we have learned about fighting economic crises over the past 80 years might actually be wrong. And even though the cherry picking of “studies” by the Romney campaign to support dubious economic arguments has become blatantly obvious, little of the public seems to be taking note. It is not about competing economic analysis, which is a terrific thing to have, but about the transparently partisan and simply weak effort that good economists like John Taylor and even Greg Mankiw are making at defending one side of the argument. It is clear that money in elections is not only used in an attempt to buy votes, but also economists. On the bright side: at least in this market, the laws of supply and demand seem to be working nicely.