There were three posts yesterday. The first was “Further Further Thoughts On Death By Excel:”
A few somewhat unrelated notes on the Reinhart-Rogoff fiasco:
1. I thought I remembered another recent Excel disaster — and sure enough, spreadsheet coding played a role in the London Whale disaster at JP Morgan:
After subtracting the old rate from the new rate, the spreadsheet divided by their sum instead of their average, as the modeler had intended. This error likely had the effect of muting volatility by a factor of two and of lowering the VaR . . .
2. There’s going to be some back and forth about modeling strategies, data choice, and so on, and I’m pretty sure some people will try to say that R-R were basically right. At this point, however, it’s reasonably clear what the data will say, because others have created data sets that more or less match what R-R claimed to have looked at; e.g., this working paper from the OECD. Here, for example, is growth for developed countries starting in 1790:
There is a negative correlation between debt and growth in the data; we can argue about how much of this represents reverse correlation. There is not, however, any red line at 90 percent. And that red line has been crucial to R-R’s influence — without the “OMG, we’re going to cross 90 percent unless we go for austerity now now now” factor, the paper would never have had the influence it’s had.
3. It’s important to make a distinction between the R-R book “This time is different” and the paper. The paper got undeserved credibility from the book; now the book may be devalued by the paper. But they’re quite different.
The book had a sound empirical strategy: it focused only on extreme events, then described what happened around those events. Because of the severity of the shock, it was reasonable to infer that whatever happened around crises was in fact crisis-related, so problems of causation were sidestepped.
The paper didn’t do any of that — it just looked at simple correlations, without making any effort to untangle causation. It wasn’t worthy of the authors. And they behaved badly by digging in when critiques surfaced, rather than responding with a good-faith effort to sort out what was really happening.
Again, however, the larger story is the evident urge of Very Serious People to find excuses for inflicting pain.
The next post was “Horse Soldiers Blogging:”
When I was little I loved, just loved the John Ford movie The Horse Soldiers, a dramatization of Grierson’s Raid; I think I made my father take me four times. As an adult, unfortunately, I found it more or less unwatchable — the prettification of war just won’t wash for me now. (The famous scene with the military school students never happened during that raid; something like it did happen during Sherman’s sweep through the Carolinas, but it was a slaughter).
No matter — the reality was actually more dramatic than the movie, with the real Benjamin Grierson — a music teacher who hated horses — a far more interesting character than the one John Wayne played.
And now that we’re at the 150th anniversary of the raid, I’m going to have a chance to start indulging my U.S. Grant obsession. I’ve always had a thing for Grant — the stubby, unprepossessing guy who failed at most things until he turned out to have a unique understanding of modern war. The scene at Appomattox — aristocratic Lee in his dress uniform surrendering to disheveled, unimpressive-looking Grant — represents for me the coming of the modern world.
On to Vicksburg!
Update: Some readers are portraying Grant as a crude attrition-type general based on his record in Virginia. But it’s pretty clear to me at least that the problem in Virginia was basically the Army of the Potomac, or more specifically many of its officers, who were spooked by Lee and repeatedly failed to carry through even when they had clear advantages. Grant had completely outmaneuvered Lee several times, for example at Petersburg, which should have been a Vicksburg-style sweep, only to have the moment thrown away by sluggish commanders; the same thing happened to Sheridan in the Valley. And in the final chase to Appomattox, once the Army of the Potomac finally realized that victory was at hand, Grant showed every bit as much vigor as at Vicksburg.
He’s right about the Army of the Potomac. The last post of the day was “Blame The Pundits, Too:”
Still Reinhart-Rogoffing; I think it’s important to be clear that R-R aren’t the only ones at fault here. In particular, the people who cited their work don’t have the right to claim innocence, because how could they know that they were being given bad data?
The fact is that R-R was controversial right from the beginning; and very early on, although we didn’t know about the coding error, we knew that they had made a major blooper by citing the US contraction after World War II as an example of debt overhang, when it was actually just postwar demobilization. That should have made everyone suspicious from the start.
Yet the VSPs not only grabbed hold of the alleged result, they wrote again and again as if this highly disputed claim was a known fact. Thus just a few months ago the Washington Post, attacking those who wanted to reduce the focus on deficits, wrote,
If [debt projections are] even slightly off, debt-to-GDP could keep rising — and stick dangerously near the 90 percent mark that economists regard as a threat to sustainable economic growth.
Not “some economists”, let alone “some economists who have been sharply criticized by other economists with equally good credentials”, but “economists”.
This is deciding what you want to believe, finding someone who tells you what you want to hear, and pretending that there are no other voices. It’s deeply irresponsible — and you can’t blame Reinhart-Rogoff for that mistake.