Thursday, 4 October 2012

Was the financial crisis the fault of DSGE modelling?

I am, like many, in awe of Bank of England director and economist Andy Haldane. However I did wince a bit at his recent Vox piece. He looks at the extent economists are to blame for the financial crisis, and he makes two interrelated claims. Having noted that central banks have traditionally been concerned with the “interplay of bank money and credit and the wider economy”, he suggests that this changed in the decade or so before the crisis. He then says

“Two developments – one academic, one policy-related – appear to have been responsible for this surprising memory loss. The first was the emergence of micro-founded dynamic stochastic general equilibrium (DGSE) models in economics. Because these models were built on real-business-cycle foundations, financial factors (asset prices, money and credit) played distinctly second fiddle, if they played a role at all. 
The second was an accompanying neglect for aggregate money and credit conditions in the construction of public policy frameworks. Inflation targeting assumed primacy as a monetary policy framework, with little role for commercial banks' balance sheets as either an end or an intermediate objective. And regulation of financial firms was in many cases taken out of the hands of central banks and delegated to separate supervisory agencies with an institution-specific, non-monetary focus.”

There is obviously some truth in this, but are these really major factors behind the financial crisis? Imagine looking at the following chart in 2005 or 2006. The increase in leverage that began in 2000 is both dramatic and unprecedented. (Much the same is true for the US.) Was this ignored because central bankers said this variable is not in their DSGE models? In my experience those involved in monetary policy look at a vast amount of information, particularly on the financial side, even though none of it appears in standard DSGE models, and even though their ultimate target might be inflation. For some reason monetary policy makers discounted the risks this explosion in leverage posed, or felt for some reason unable to warn others about it, but I very much doubt these reasons had anything to do with DSGE models.

UK Bank Leverage. Source: Bank of England Financial Stability Report June 2012
I say this because to place too much weight on the culpability of DSGE models and inflation targeting can lead to overreaction, and may sideline more fundamental issues. (I don’t, by the way, think Haldane himself falls into either trap: see this interview for example.) Let me take overreaction first. It is one thing to claim, as I have, that the microfoundations approach embodied in DSGE models encouraged macroeconomists to avoid modelling difficult (from that perspective) issues like the role of financial institutions in credit provision. It is quite another to suggest, as some do, that DSGE models are incapable of doing this. This second claim was false before the crisis (e.g. Bernanke, Gertler & Gilchrist, 1999), and has clearly been shown to be false by the post crisis explosion of DSGE work on financial frictions. Forming a rough consensus around a reasonably simple and tractable model of the crisis that can also assess the subsequent policy response will not happen overnight (it never does), and I suspect it will involve tricks which microfoundation purists will complain about, but I’m pretty certain it will happen.

Andy Haldane talks about the need to model the interconnections (networks) of actors and institutions in order to understand how sudden crises can emerge. This must be right, and recent work[1] that begins to do this looks very interesting. However what seems to me critical in avoiding future crises is to understand why leverage increased (and was allowed to increase) in the first place, rather than the specifics of how it unravelled. As I suggested here, we may find more revealing answers by thinking about the political economy of how banks influenced regulations and regulators, rather than by thinking about the dynamics of networks. We should also look at the incentives within banks, and why short term behaviour in the financial sector may be increasing, as Haldane himself has suggested. Investigating networks is clearly interesting, important and should be pursued, but other avenues involving perhaps more conventional economics and political economy may turn out to be at least as informative in understanding how the crisis was allowed to develop.


I wrote this before reading this from Diane Coyle, which is well worth reading if you think all microeconomists must be in favour of DSGE. We both went to the conference she mentions, and my own rather different reactions to it partly inspired my post. I'd like to say more about this quite soon.

[1] See, for example P Gai, A Haldane and S Kapadia, Journal of Monetary Economics, Vol. 58, Issue 5, pages 453-470, 2011, and other recent work by Kapadia.


  1. Saying that the financial crisis was the fault of DSGE modeling sounds to me like saying the financial crisis was due to algebraic modeling of economies. A tool cannot be at fault for anything, the problem is macroeconomists.

    I was writing a list of things that I think are at the heart of the problem, but your linked article from Diane Coyle summarizes it pretty well:

    My view is that if macroeconomics does not abandon its obsession with being able to write down analytical 3 equation models with maybe 12 or even 20 variables to explain and predict what is happening at large scale in the economy, it will lose all meaning and purpose.

    Physicists cannot easily solve the three-body problem and macroeconomists pretend to analytically derive equations for whole economies? The intellectual arrogance of economists is too damn high.

    1. The key distinction I would make is this. Writing down and analysing small models (microfounded or not) is very useful in helping to understand what systems can do. It would be crazy not to do this. However this is not the only thing macroeconomists should do. It is also very useful to see what sort of relationships match the data.

    2. Yes, I agree. Which is why I wrote that the problem is not DSGE modeling. Note that DSGE modeling is not the only way of modeling microfounded models, and "forcing" equilibrium may not be appropriate in all situations.

      I would also add to the mix of things to do dynamic simulations. It seems to me that agent-based computational economics should be a very fruitful field. Economies are systems ridden with complex feedback loops, it seems implausible to me that a good model will be "solvable". This means that, while simplifying might be a good exercise in some cases, it is very likely that the models one is able to solve are too simple to realistically match the real world.

    3. Macroeconomists do not pretend to analytically derive equations for whole economies. Contrary to what Diane Coyle wrote, even small scale DSGE models cannot be solved analytically. DSGE models do capture feedback loops - they're general equilibrium models.

      That said, I do agree with Simon that small analytical models can be very helpful in understanding things. Paul Krugman has written about this too. His use of very basic IS/LM to make predictions in a liquidity trap is an excellent example of this type of analysis in practice.

  2. “Hicks rejection of IS/LM by Lars P Syll
    with background story in a comment from Paul Davidson. and

  3. derrida derider4 October 2012 23:37

    Shorter Simon: "DSGE is not to blame because the people responsible for the real world knew it was useless". As one of those unfashionable Popperians I'm not sure I'd care to "defend" my preferred modelling framework in those terms.

    Seriously, Simon, if the only way we can rescue DSGE is to load it with non-micro founded auxiliary hypotheses, all calibrated to fit the most recent crisis, then have we gained from the whole approach beyond some rather kludgy maths? We'd have done better sticking to old-style MEMs while keeping the Lucas critique as a footnote "Warning: Contents are hazardous to small children. Read safety instructions carefully. Do not overload; due to micro-scale defects model may collapse under load".

    After all by your own description that's what monetary authorities pretty well did in a highly informal, and therefore even less reliable, way.

    1. In the spectrum between 'we focus on DSGE models' to 'all DSGE models are useless', there is an optimum that is no where near either extreme. I find looking at DSGE models, with or without dubious microfoundations, useful in helping my thinking about how economic systems can work.

  4. "For some reason monetary policy makers discounted the risks this explosion in leverage posed... I very much doubt these reasons had anything to do with DSGE models."

    I think the best way to read Haldane is that he's trying to explain that reason. He claims it has something to do with the models that macro denizens used (predominantly DSGE) but isn't specific about what. Okay... is that plausible?

    I think it is. Folks were convinced that microfounded DSGE models dominated non-microfounded models (which might possibly be true for some definition of dominance), and that "therefore" the DSGE models we had were superior to older models so that older models could be effectively ignored. That doesn't follow.

    Macro research was way too focused on what could be modeled with DSGE, and way too little focused on what needed to be modeled. That's what Haldane is saying. And he's right.

    1. I agree about the focus on macro research, but I think there is much more to why policy makers discounted leverage risk.

  5. derrida derider4 October 2012 23:53

    AS a PS, that bit about "... thinking about the political economy of how banks influenced regulations and regulators" is a good example of how preoccupation with the wrong questions is actively harmful because it makes you forget stuff you already knew about the right questions.

    It is an absolute commonplace, dating back to Bagehot at least, that the first people swept up in the enthusiasm of a boom are the regulators; endogenous regulation is surely a no-brainer to put in any macro finance model claiming micro-foundations. But it's not a natural fit for a DSGE - so much the worse for DSGE.

  6. "There is obviously some truth in this, but are these really major factors behind the financial crisis?" To be precise, Haldane does not claim in his piece that the two developments are the major factors behind the financial crisis.

    1. True, but it could be interpreted as such. I hope what I wrote made that distinction - see my penultimate paragraph.

  7. I posted a long comment on Diane Coyle's blog. I won't repeat it here, but I'd make the short point that if the centre of the crisis was some kind of asset bubble, there was a huge culture/ideology problem with mainstream macro which caused a lot of economists to either deny the existence of bubbles, or to treat them as a problem which could not be dealt with. (How do you know it's a bubble?) I feel like DGSE may have played a part in this kind of attitude.

  8. It's my impression that economists worry greatly about government debt, but not private sector debt. The assumption being that individuals entering into a voluntary exchange will look out for their own interests.

    And for this reason, I believe economists did not worry too much about the growth of private sector debt and even now, when they look back at the crisis, many are attempting to re-interpret it as a sovereign debt problem.

    That at least suggests to me a big blind spot, and the blame would fall on GEQ models more generally, due to the optimality results that flow from them.

    So yes, IMO, the failure of economists to sound the alarm bells is due to DSGE.

  9. You will be notified about this topic really well. I, you, as anyone can easily understand, I really like how to have put the light on this topic.

    Maria Rivero

  10. Three points.

    First, the chart originally came, I believe, from the report of the Independent Commission on Banking (of which I was a member). It is extraordinary that nobody seems to have been aware of this before the crisis.

    Second and far more important, the dominant feature of the pre-crisis economy was a massive increase in leverage. It is surely not surprising that such an increase is matched by a huge increase of leverage in the sector that intermediated almost all this debt, namely, the banking sector. So the rise in leverage in the economy should have given a warning about what was also happening to finance. But none of this seems to have been present in pre-crisis macroeconomic thinking. So surely it was a fundamental failure of macroeconomics to incorporate the financial sector in its thinking - a mistake that a pupil of, say, Minsky, would not have made. Whether this mistake has now been rectified I do not know.

    Third, I agree that the biggest mistake was in conventional financial economics. The view came to be accepted that the financial sector could not make really big collective mistakes. That was absurd, we now know.

  11. Martin Wolf was kind enough to reply to a mail which I sent him back in August 2008 commenting on a piece he wrote in the FT which had irritated me. Mr. Wolf told me that the world was doing fine, all was well, imperialism is a normal state of affairs (US financial and military imperialism in this case) and that there was no reason to be concerned.
    4 Weeks later the world's financial system collapsed.
    Given that Mr. Wolf was/is a professional shill for the financial system (the FT is logically a propaganda source for big finance) and was a part of the problem, I personally find his "epiphany" somewhat hard to take seriously.

  12. As far as the DSGE model goes, it is amazing that anyone still takes it seriously. The idea that such a complex dynamic system as the economy with multiple seriously nonlinear (+ve) feedback loops can be thought of as having a natural tendency to equilibrium is just absurd.

    Why an Oxford economists is still thinking of the world in this way defeats me. Just pop up the parks road to the Department of Engineering Science and ask someone to explain control systems theory to you. Nonlinear systems can only be analysed if they are "linearised" at some local operating point and then only allowing small perturbations about this point - large changes will cause your model to become invalid and force a re-lineraisation.

    When a model (DSGE) is clear utterly useless at predicting anything other than anomalous steady state conditions then surely it is time to throw it away?

    This is more like a real model of a dynamic economy:

    Why don't economists talk to engineers?