There’s lots to find encouraging about today’s One Bank Research Agenda Launch Party. [Caveat: my recovering kidney meant I could not sit for very long on the mis-shaped BoE conference centre chairs and had to bail out of most of it].
1. It symbolises the new emphasis placed on research. Along with the reorganisation of that function, Bank-wide, under the Chief Economist, Andy Haldane.
2. The launch falls in with a pattern of ever-increasing openness and transparency at the Bank. The very fact that the agenda is disclosed up front, inviting comment, in a way that wasn’t done previously, allowing an informal holding to account for how taxes used to fund it is spent, has to be applauded. [We trust there will be a ‘where have we got to?’ Party to at some point.]
3. The stated openness to new ideas, new methods, new thinking, and the recognition that what was thought before may not have been the answer. It’s not long ago that this would have been unheard of (in public) in a central bank, when it was held that the credibility of all functions depended on communicating omnipotence, and communicating as little as possible.
4. The ‘crowdsourcing’ initiative. Although the Old Street tech Roundabout language made me cringe a little, the idea is helpful. Write down lots of important policy questions that others in academia might not be aware of, and try to encourage others to help. Those interested in macro and finance in academia are strewn about the country, isolated in small groups. Initiatives like this help bring people together, not only physically, but in instilling common purpose.
But. Everywhere I felt that babies were being thrown out with the post crisis bath water.
1. We had the suggestion that before the New Enlightened Era, which began some time after Northern Rock collapsed (?), researchers de-emphasised empirical work, and did not do inductive research – taking data to the theory, rather than the other way round. Well, that was news to me. Perhaps around the mid 1990s I started to hear the term ’empirical macro’. But I know now that it much predated when I first understood it, and goes back to work by Sims, Sargent, Quah, Geweke, and was developed later by Stock, Watson, Reichlin and others. This sub-discipline was all about putting the data first. And, incidentally, using a lot of it. Although this would probably still be called, insightfully, ‘small data’, in today’s lexicon.
2. There was the implication that we should stop looking for causation, and satisfy ourselves with correlation. Relatedly, theory [unless it’s ‘complex systems theory’] is out. But lurking behind the tension between empirical and theoretical macro is the Lucas Critique. The upshot of which is that correlations don’t provide reliable guides to policy. Neither does theory, mind you. But nothing about the crisis explains how correlation can now suffice.
3. There’s a contradiction between the pervasive idea that old methods and lessons are done for, and, well, almost everything else the Bank is seeking to do outside its Agenda. For example, the inflation target; the pursuit of it by the setting of interest rates and quantitative easing; the credit easing policy and the calibration of the subsidy; the support for higher capital requirements; macro prudential intervention in respect of high loan to value mortgages. All these things are underpinned by decades of theory. It seems that we aren’t sweeping all that away today. But if not, why not? Are some of these bits of wisdom still reliable? All the while received understanding is scorned, a bunch of hard-working staff are turning the handle on the Bank’s COMPASS model, producing a forecast and studying the impact of alternative policies. How does that square? I suggest that these old-fashioned actions speak louder than the radical, landscape altering rhetoric, which I suggest is impractical and overblown.
4. There are some dangers in the new research agenda. Read one way, it purports to set the Bank on a path to research everything, using every possible method, quantitatively and quantitatively, exploring all possible interdisciplinary synergies. One of the very few management lessons that ever made it through my muddled business consciousness is that if you set out to do everything, you end up doing nothing. Despite the ambition in the program, discussants, amusingly, managed to find all kinds of gaps and nuances that were missed. Bank staff were diligently writing these suggestions down, but I hope most of them are set aside, as complicating an already difficult task.
5. I’m in two minds whether all the nice infrastructure of Themes and Questions and the Agenda is catalysing and energising, or a necessary evil to show that due process is being adhered to, or a terribly costly, suffocating superstructure. I can only imagine all the meetings and brainstorms that gave birth to it all. Most of the successful research that appears in the top journals [granted that might not be what the BoE is after] was done by individuals operating without the benefits that all this management overhead conferred. They were researchers being paid, choosing their own topic, finishing to their own timetable, responding to incentives. An opposing model to the one being followed is: these are our core purposes: we think that research is a necessary part of them; and this is how attractive the job will be if you come and work for us [fill in pay and conditions here].
6. Today was a real showcase for various strands of what you might call heterodox economic thought and practice. Or not. Two things bugged me about the discourse. First, there were so many instances throughout the short time I was there where the determination was to do heterodox thing X, when, actually, this has been part of standard macro practice [eg empiricism]. Second, was the way the heterodox discourse drew common cause in every possible critique of mainstream practice, even though these critiques are sometimes mutually exclusive. It reminds me of listening to some who espouse alternative medicine. In fact several, mutually exclusive alternative medicines.
7. I still am not getting Big Data. Today’s slide show was delivered like many I have seen. Shouted, with missionary zeal, as though The Fog Is To Be Lifted From Our Unseeing Eyes. Why doesn’t someone simply say: computing power and the internet means there’s a lot of data now, and you econometricians and forecasters can do a lot more of what you used to do, but with the same health warnings? There’s also an element of ‘you mainstreamers have really missed something here’. Yet heterogeneous agent macro and micro people have steadily been chewing through many huge data sets. Big Data – or how about abundant data and computing power – is not a methodological challenge to mainstream economics, it’s an opportunity for it, and one already being embraced.
8. The Bank of England tweeted out, graciously, Charles Goodhart’s comment that people will be astonished in the future to learn that central bank models exclude the banking sector. Minor detail – so does the BoE’s model, still, some years into our post banking crisis phase! As the BoE tweeter hinted subsequently, there are models with banks in the ‘suite’ of models that make up their forecasting toolkit. But is this a substitute for the real thing, building them in the core model? I doubt it, though the argument is more nuanced, and not as ridiculous as it seems on the surface.
If you haven’t already bored of my thoughts on this, I have blogged here and here on related issues.
longandvariable 
Big data is the new bacon. Or maybe kale.