The models didn’t work then, and they don’t work now

Academic economists aren’t always the best when it comes to precise market timing. Interested to learn the lessons of the Great Inflation of the late 1970s, the best and brightest minds gathered at a place called Woodstock (there’s more than one, didn’t you know) to see what they could agree on. But it had taken them thirty years to reach this point in their investigations, which meant it was September 2008 by the time they met. Two weeks earlier, Lehman Brothers had gone into administration. Markets were in free fall, and it wasn’t a Great Inflation but a Great Deflation that was concentrating minds. Some of the Central Bankers scheduled to attend couldn’t even make it, in the end. Another five years passed before a (very useful) book would come out to mark the event.

With the Global Financial Crisis (GFC) generally reckoned as having taken place from 2007-2009, this year marks the mid-point of its the tenth anniversary. Last month, former US policy-makers Ben Bernanke, Hank Paulson, and Tim Geithner were reminiscing about it at a Brookings press conference, including a few spared thoughts voicing “some concern about the next crisis”. It’s just a pity they couldn’t voice any concern about the next crisis before the last one, I thought to myself.

But the academic economists have been at it, too, with respect to their GFC retrospectives. The latest one to cross my desk was in the Summer 2018 issue of the Journal of Economic Perspectives (JEP), called “Macroeconomics a decade after the Great Recession”. JEP is a fantastic resource for economists. And it’s free. Its publishers say it “fills the gap between the general interest press and academic economics journals”. Well, if they mean the rest of us might understand 20% of what they’re writing in the JEP – rather than the customary 2% with respect to standard academic articlea – then I suppose they do have a point. Either that or the economics profession is seriously over-estimating its teaching abilities, and/or its students (including me).

It was the article by Professor Jordi Galí that first caught my eye: “The state of New Keynesian Economics: a partial assessment”. It caught my eye because this was from one of the authors of one of the first big papers (in economics circles) on New Keynesian economics, published in 1999. And, as it happens, this was a paper I had to review and explain to my classmates during the course of my MSc studies.

But before we get to Professor Galí’s enlightening summary, it might be useful to have a brief history lesson (of economics), to see how he fits in. As capitalism flickered into modern life, it was Adam Smith who first noticed the importance of changing business practices (he called it the Division of Labour) and the benefits of leaving entrepreneurs to go about their business unfettered (thereby benefiting from the Invisible Hand of the free market). There then followed something of a division of labour craze (significant parts of which were linked to the first Globalisation boom), but with a significant technology boom thrown in (canals and railroads, anyone?), even though one of the by-products of this seemed to be quite volatile financial markets along the way. But where markets saw volatility, economists saw signs of a first Great Moderation, including because of a new US incarnation known as the Federal Reserve. Things reached a head (not the permanently high plateau Irving Fisher envisaged, alas) in the Great Crash of 1929, and subsequent Great Depression. This spurred Lord Keynes into suggesting that a much more Visible Hand of government intervention was called for, in the form what we know today as fiscal policy. But few fellow economists understood exactly what Keynes’ highfalutin theories were really founded on; many thought him a bit of a grand-stander. And, anyway, about then began another long wave of globalisation after World War II (the Marshall Plan helped), with new and booming technology to boot, and renewed attempts at rationalising the possibility of a Great Moderation of the dreaded business cycle, even though financial markets – not for the first time – were starting to become rather volatile again. The Phillips curve was one such attempt at calibrating economic performance, with Professors Samuelson and Solow proposing an actual tabular trade-off between inflation and growth. When this, too, came to grief in the aforementioned Great Inflation of the 1970s, it fell to Milton Friedman and others of a similar ilk to suggest that these Central Banks contraptions ought to be put back to sleep, with computers doing the work of issuing money instead.

All of which brings us back to Professor Galí and his, three-equation New Keynesian perpetual motion machine. Central Bankers didn’t want to be put back to sleep as Professor Friedman was suggesting, and Professor Galì’s new machine gave them something positive to do. You think I’m kidding about the perpetual motion bit, right? But I’m not. If inflation goes up, the Central Bank raises interest rates by more (in equation 1), and that causes demand growth to slow (in equation 2), inflation to come back down (in equation 3), and the business cycle to go on (back to equation 1, and repeat). It’s more or less that simple, believe it or not.

And it really is perpetual motion – in theory – unless prices happen to fall. But when Professor Galí and his colleagues were dreaming up the New Keynesian model in the late 1990s, the idea of falling prices was the last thing on their minds given what had happened in the prior thirty years (plus, remember that the Great Inflation conference was still a decade into the future).

Central bankers picked up on the New Keynesian super-computer. Indeed, almost nobody – the authors aside – appears more in the bibliography of the original 1999 paper (I’ve counted them, by the way) than Ben Bernanke, then a Fed Governor, but later to be its Chairman. New Keynesian economics by the early 2000s wasn’t just an academic endeavour: it’s conclusions had reached the heart of policy-making circles.

But, just like the Titanic on its maiden voyage, the New Keynesian machine ran into its nemesis – the threat of falling prices – almost on its first voyage, in the aftermath of the 2001 recession. Interest rates couldn’t go negative, reasoned the boffins, but at the very least, interest rates should be set a close as possible to zero, and to be credibly kept there, so as to keep the risk down (this happened in August 2003, by the way).

With interest rates near-zero and staying near zero, policy-makers duly avoided what they thought was the iceberg of falling prices, even though it wasn’t an iceberg at all (while available prices did fall in the early 1930s, they were the prices of base metal and commodities, not the services-heavy CPI basket of the post Dotcom era). But, having steered the MS Global Economy so violently to miss the (non-existent) iceberg, policy-makers ended up capsizing it instead, swamped as it had become with almost free money, available to almost anyone.

Unlike the Great Depression, where policy-makers didn’t ease policy enough, the Great Recession occurred because policy-makers had eased policy too much.

But this is what Professor Galí had to say about it all in his Symposium piece (remember now we’re moving back in the world of drier academic prose): “The standard New Keynesian framework as it existed a decade ago has faced challenges in the aftermath of the financial crisis 2007-2009”. Or in my English: the New Keynesian models failed to predict the Great Recession. In my view, the models had done far worse: they’d help cause it.

But it was the next paragraph that really had me on the floor (not in laughter: I’d just fallen off my chair): “However, none of the extensions of the New Keynesian model proposed in recent years seem to capture an important aspect of most financial crises”. Or in English: the New Keynesian models are still no good when it comes to an ability to see beyond the close of business of the last data point used (or the end of last quarter, if the models are being populated with quarterly data).

Guess what came next? Take a breath, now: “But in the meantime, New Keynesian economics is alive and well”, said Professor Galí, in closing.

Or in English, to close this article myself, and as Mark Twain once said: “never let the truth get in the way of a good story”.

Now, about that permanently high plateau.