A lot of intriguing debates have occurred in the last month or so that merit individual blogs, but markets have been preoccupying me, so I have been unable to write at length. In some ways, these four subjects overlap:
1. Andy Haldane is one of the world’s most interesting central bankers. But this speech is ultimately disappointing. The summary: people are myopic. This is a central tenet of behavioural economics. It is also implicit in mainstream economics because most models of inter-temporal choice use a non-rational time discount factor. Haldane is the latest in a line of commentators to argue that stock market investors have short time horizons. But central bankers shouldn’t use stock markets as an example, use central banking. Only someone with an incredibly short time horizon thinks it matters whether the Federal Reserve changes interest rates in September by 25bps. The most encouraging aspect of Haldane’s foray unto the current debate about corporate investment spending is that it inspired a brilliant riposte from Chris Dillow. Chris makes one of those rare observations which once you’ve heard it you realise that it’s profound, obviously true, and original.
I’ll try to summarise, with a bit of background. Eugene Fama and Robert Shiller shared the 2013 nobel prize in economics. Many commentators argue that they hold opposite views: one believes in efficient markets, the other in behavioural psychology. But they actually both agree on the main empirical facts: value investing works – i.e. yields predict returns. (And by the way, so does John Cochrane). Fama says ‘cheap’ assets must be compensation for risk, Shiller says people’s beliefs are dominated by behavioural factors. But empirically, who cares – the facts are clear. Dillow’s profound point, as I interpret it, is that the same facts refute Haldane’s arguments. If value stocks deliver higher returns than growth stocks (and growth stocks have higher rates of investment), markets have a ‘growth bias’ – investors actually overpay for growth. Now this does not refute the truism that humans are myopic. Myopia can encourage trend-following and momentum investing. Currently, for example, investors are putting a far higher rating on Amazon shares – a company that has an extremely high rate of investment spending. Simultaneously, IBM – a company which is doing very little investing, buying back shares and paying dividends, something Haldane bemoans – has performed very poorly, and has been steadily derated. (Ironically, Warren Buffett – who is definitively NOT myopic – is the biggest shareholder in IBM). So … markets, and humans, are more complex. Stock market (and human) myopia does not imply “short-termism” in investment spending. It can imply the opposite.
2. Fama and Shiller agree on facts and disagree on theory – this leads nicely onto ‘mathiness’. Paul Romer has stirred up a great debate. I think the essence of his argument is an identification of political bias in economics. He is a bit subtler than this, arguing that mathematics is being used to obfuscate. He is right on both counts. The obfuscation point is more general, the politicisation less interesting, because it’s so obvious. Ask yourself this: will Robert Barro ever produce an empirical paper showing a very significant multiplier effect? Enough said.
The tangents of the debate Romer has started have been more rewarding, in particular he has alerted us to this set of papers from a Boston Fed conference on the Philips curve in 1978. Two observations. First, one of the biggest failings of economics is its treatment of expectations. Eclecticism is surely right: it is clear that humans can be rational, adaptive, irrational, and non-rational – it all depends on the circumstances and who they are. Keynes got this, and very little relevant progress (in economics) has been made since (the neglected Mordecai Kurz is an exception – HT Woody Brock). Franco Modigliani makes the wisest contribution.
The second substantive point to come from Romer is about macro-economic modelling and forecasting. I am not an econometrician. I briefly tried macro-modelling and forecasting about 20-years ago and concluded that it didn’t help me. Which is not to say it isn’t a fascinating intellectual exercise – and Ray Fair and Simon Wren-Lewis illustrate why. But as a practitioner focused on asset prices, what I really care about is the probability of recession, and models are much worse than the back of an envelope and a handful of financial market indicators (if you want to know which, read the FOMC transcripts from 2007 and 2008 – Ben Bernanke agrees with me!).
3. This leads me to China. The noise has been deafening, with the notable exception of Martin Sandbu and colleagues, at the FT. How Sandbu can sustain this level of quality on a daily basis is remarkable. In thinking about China, let’s start with some facts. China may already be in recession (actual contracting GDP) or it may be close to recession. NO econometric model predicted or will predict this in a timely fashion. But the declining PE multiple of the Chinese stock market over the last 5-years (until last Summer) has consistently implied a rising probability of recession (spot GDP growth forecasts are silly). The recent crash in Chinese stocks should not be dismissed lightly. Keynes wrote more relevantly on the probability of recessions than any economic theorist I’ve read in the last 50-years. “Notes on the Trade Cycle” is a chapter of genius in the General Theory. To paraphrase one key insight: recoveries are gradual, recessions are abrupt. To understand what’s happening in China better, I have found more insight in Hicks’s little book on the Trade Cycle – which is all about the accelerator – than any forecasts from Chinese economists. I started re-reading Hicks because I was sceptical of Paul Romer’s side remark about economics pre-Samuelson being devoid of mathematical rigor. Hicks was never guilty of ‘mathiness’, but he was rigorous, and insightful.
4. Finally, Corbynomics. I’ve already written a blog on this. And I have rightly been rebuked for not taking his PQE more seriously. I have one observation: it matters who is in charge of the printing press. Even very smart people don’t seem to realise that institutional “independence” is not binary – it is a matter of degree and it involves a division of labour. Is the judiciary ‘independent’? Well, yes and no. They don’t make the laws (although sometimes precedent and interpretation comes close to law-making) – parliament or the legislature does – or in the case of constitutional laws, referenda are required for changes. So in a democracy, courts do not have some absolute ‘independence’. But the decisions of courts should be made independently of any politicians. This is uncontentious in political theory and practice.
So too in central banking. Parliament should determine the objective of the central bank – and the objectives and tools should be relatively constant, because rules are important. Interest rates and money printing should not be subject to electoral cycles, in the same way that constitutional law shouldn’t be. There is a very strong consensus that inflation should be kept relatively stable, and that central banks should control the printing press, subject to a set of mandated objectives. Now, a number of economists from very different traditions, including myself, John Muellbauer, Simon Wren-Lewis, Mark Blyth, Steve Keen, Frances Coppola, Anatole Kaletsky, and many others have argued that the Bank of England (and the ECB), should be given the contingency power to make cash payments to the household sector to prevent deflation and recession. Simon, Mark and I also argue that the Bank should remain operationally independent, and we think there are simple, prudent ways of addressing the technical implications for the Bank’s balance sheet. I suspect we also all have no objections to sensible state-led infrastructure spending. The key point is that the case for keeping these policies institutionally separate is compelling. Infrastructure should be determined by the need for infrastructure, monetary policy should be effective counter-cyclical demand-management subject to an inflation target. In fairness to Corbyn, I think Adair Turner and Larry Summers are equally unclear on the importance of these distinctions – so he is in good company. Other than that, I think what Corbyn reveals is that a significant part of the electorate is crying out for interesting and important new ideas. I just wish the genuine policy innovations which are out there were being proposed.