Saturday, November 3, 2018

An information equilibrium history of the Great Recession

I mentioned at the beginning of this year on my book website that I was thinking about writing another book about the macro history of the US as told through dynamic information equilibrium and the resulting economic seismograms. I've been collecting the various models on this blog to put them together into graphics that tell at least one version of history. Previously, I've given evidence that women entering the workforce leads nearly every other measure of growth and inflation in the 70s and 80s. Lately, I've been working on the Great Recession. Here's the seismogram (click to enlarge):

Red-orange indicates negative (i.e. bad) shocks, while blue indicates positive (i.e. good) shocks (rising unemployment is "bad", but rising income is "good"). The labels are identified in footnote [1].

While much of the focus of commentary about the recession was on the Lehman collapse and the Fed meetings immediately preceding it (along with the fall in the stock markets as measured by the S&P 500), these actually come in the middle of the recession process . The first thing that happens by far is the drop in hires in construction (labeled "HIR 2300" based on the JOLTS code) in mid-2006. Around that time, Paul Krugman (e.g.) was talking about a housing bubble deflating (he had been forecasting it earlier in mid-2005) [2]. The shock to housing starts (HS) doesn't come until later (though the shock to starts occurs over a longer period, you can see that hires begin to decline just before housing starts begin to decline).  The drop in construction hires also comes right before the halt in the Fed rate increases that had started in 2004.

Before the NBER-defined recession gets underway, there's a drop in conceptions (per this NBER working paper) that's roughly coincident with (but genuinely followed by) two Fed conference calls in 2007 about the financial markets reeling in the collapsing housing bubble (the negative shock to the Case Shiller index) as well as the first Fed rate cut. The rest of the stuff that is associated with a recession in the media (stock market drops, GDP declining, unemployment rate rising) all come much later during the NBER-defined recession.

Personal income (PI) continues to climb ahead of its typical pace through most of 2007, and wage growth continues to increase (i.e. accelerate) almost until the NBER recession end.

While I've heard many stories about excessive debt being a cause behind the Great Recession, most of the negative shocks to debt measures come later (i.e. debt became a problem because of the recession). Although not shown in this graph, consumer credit takes a hit only as the NBER recession is ending. This is not to say that debt levels didn't contribute to the size of the recession (i.e. making it worse), but rather that they didn't contribute to its timing (i.e causality).

Any causality analysis would put construction hires at the beginning of the story, but oddly the shock to construction job openings comes along with the rest recession — barely leading the shock to job openings of all kinds. In fact, there's a surge in openings around the same time. It's the largest difference in timing for all the JOLTS sectors. That is to say jobs were still being advertised in 2006 (until 2008), just fewer were being hired. This doesn't indicate a pessimism about the housing market (which seems like it would show a fall in openings), but rather a labor shortage of some kind. Were employers unwilling to raise wages? Unemployment had reached its lowest level since before the 2001 recession, so maybe there was a genuine shortage of workers.

Was it xenophobia?

I am going to offer a speculative answer that I do not think I have ever seen offered as a possible reason for the Great Recession: xenophobia. There were a series of protests from March of 2006 to against anti-immigrant legislation being introduced (some of which passed, and in various jurisdictions E-verify was mandated in 2006 to prevent employers from hiring undocumented workers). The shock to construction hires begins right around the same time as those March protests, and every year since 2004 saw a decrease in immigration from Mexico:

The linked article doesn't get this causality right:
Immigration from Mexico dropped after the U.S. housing market (and construction employment) collapsed in 2006. By 2007, gross inflows from Mexico dipped to 280,000; they continued to fall to 150,000 in 2009 and were even lower in 2010.
According to their data, immigration started dropping before 2006 (the peak is in 2004), but given noise in the data and the annual temporal resolution the best we can say is that construction employment and immigration from Mexico dropped approximately concurrently.

I have written before on how much of an effect a drop of 2 million people in the labor force due to immigration restrictions would cause — about 1 trillion dollars in NGDP. Assuming a linear trend past 2007 in the increase in just undocumented immigrants (using Pew data), by 2009 there were 1.8 million fewer undocumented immigrants (11.3 million) than would be expected by the trend (13.1 million). While there would need to be more detail added (accounting for the decline in documented immigration as well as fraction of those two populations in the labor force), this gives us an order of magnitude that is not trivial compared to the size of the Great Recession.

Again, this is speculative. However it is not implausible that the anti-immigrant sentiment of the mid-2000s ended the "housing bubble". Employers continued to look for workers in construction, but suddenly were unable to hire as many starting in 2006 due to declining immigration. The worst hit states in the housing crash were California, Arizona, Nevada, and Florida — the first three being major destinations for documented and undocumented immigrants from Mexico. Since even undocumented immigrants spend money at the same grocery stores you do, sales decline. Declining construction hires is followed by fewer housing starts, and when a new family can't find a bigger house with more rooms they'll not only delay having children but opt to hold off on that house. Housing prices decline from their peak, but by now the general economic outlook is mediocre enough that the Fed starts to lower interest rates in 2007. Pessimism sets in along with the rest of the recession and a financial crisis that goes global. 


Update 6 November 2018

A correspondent sent me a link to some work by Kevin Erdmann about how there was actually an under-supply of housing going into the 2008 recession. Now Erdmann is writing for Mercatus which generally means there is a possibility of an ideological slant or at least a particular view of how economies work. Here, that reasoning is an attempt to say there was no housing bubble because there was a "fundamental" reason (short supply). But then, there was a limited supply of tulip bulbs as well. If there was no housing bubble, then it's arguable that the Fed had unnecessarily tight monetary policy (i.e. the desired conclusion in this case). Seeing as monetary policy tends to lag other measures, it's probably not the cause (but may e.g. contribute to the broader conditions and the depth of the recession).

I also want to emphasize that it is almost entirely unlikely the shock to construction hires was the only causal factor. I see it more as a trigger or a straw that broke the camel's back — in an environment of higher interest rates and general pressure from policymakers to cool the housing market, a sudden shock to labor supply makes that "cooling" suddenly look worse in a way that could change one's outlook. In the information equilibrium approach, it's sudden coordinated action (e.g. panicking) causing agents to cluster in the state space that causes recessions. Sometimes that coordinating signal is the Fed, but it could easily be shock to labor supply due to an unwarranted immigration freak out.



[1] The labels are:

HIR 2300: JOLTS hires, construction (JTS2300HIR)
HS: Housing Starts (HOUST)
C/F: Conceptions/fertility
Case Shiller: Case Shiller housing price index (also here)
HIR: JOLTS hires
HIR-ext: Extended JOLTS hires data
JOR 2300: JOLTS job openings, construction (JTS2300JOR)
Debt growth: Growth of debt (All Sectors; Debt Securities and Loans; Liability, Level)
JOR: JOLTS Job opening rate
JOR Barnichon: Job openings in data from Barnichon (2010) [pdf]
QUR: JOLTS quits
SP500: S&P 500
U: U3 unemployment rate
PI: Personal income
PCE: Personal consumption expenditures
NGDP: Nominal Gross Domestic Product
W: Wage growth (Atlanta Fed)
Debt to GDP: Ratio of previous debt measure to NGDP
CLF: Civilian labor force (CLF16OV)

The arrows on the top of the diagram indicate the two Fed meetings (black arrows) prior to the Lehman collapse (red arrow). The arrows on the bottom of the diagram show the first Fed rate increase since the 2001 recession, the beginning of the period of steady rates (mid-2006 to mid-2007) as well as the first rate cut going into the 2008 recession.

[2] I'm not tying to make any point here about "who saw the crisis coming" — only citing some news that I remembered from the time for context.


  1. I am grateful to you (and for the power of the Internet) whenever you share new work on this blog. It is truly outstanding work, as always. I have been wondering, though, whether you could strengthen the evidence for several of your hypotheses by doing a bit more work on the effects of monetary policy on the macroeconomy. The object would be to show why we can reject monetary policy as the causal factor in certain economic events (e.g., the 70s inflation, the housing collapse of 2007-08) in favor of other factors you put forth.

    With respect to your hypothesis regarding women entering the workforce as leading and therefore potentially causing the inflation of the 60s, 70s, and into the 80s, I was struck going through old Wells Fargo Annual Reports ( at how different the financial system seemed to be functioning during that period compared to more modern periods of the 1990s, 2000s, and now 2010s. I am quite intrigued by how a financial system that could expand credit so rapidly was related at the time to the massive demographic transition you cite (never mind other macro shocks, e.g., in energy, that are given primacy in other accounts of the period) and how these factors together would have been related to the phenomenon of inflation. On one hand, I am quite intrigued by a thought experiment in which a large number of people are hired and begin to be paid before they can be expected to be fully productive and generating new economic output, and how this slippage might explain a period of both inflation and rising inflation expectations in an economy, but on the other hand, I wonder if such a phenomenon could only happen if abundant credit enabled the hiring of so many workers when the output of the real economy did not appear to be growing at a high enough rate to support them. *

    As for this post, you cite another leading demographic trend, this time a decline in the number of immigrants in the construction labor force, and identify it as contributing to the collapse of the housing market and the great recession, which is quite interesting and, as you note, novel as an explanation. But again, there is the more conventionally cited factor of monetary policy, which saw the Federal Reserve rapidly raising rates and presumably clamping down on credit growth in the financial system and economy in 2004-2006, ahead of the same 2006-2008 period you are studying here.

    Is it that you have already found in your studies that the kind of correlations with respect to events you have found in the demographic factors are somehow superior to the kind of correlations one would find in financial and monetary factors, especially allowing, as you are careful to do, for quality of data and timing issues? More broadly, how would you frame your thinking on balancing the benefits of a theory whose effectiveness might be difficult to test (e.g., how well does “influx of women in the workforce” predict “inflation,” when such an influx might happen exactly once in the history of a country) against the desire to make more instrumental claims such as “factor A was the primary cause of event X,” especially in cases where a phenomenon might appear to be “overdetermined” by the information we have available and our ability to process it (e.g., when each of two or more factors does a reasonably good job of leading and correlating with an economic event)?

    * Note that it would become a separate, normative debate on whether certain members of society should be expected to bear certain costs – e.g., of inflation – in order to speed up the attainment of a new state of affairs – e.g., a fuller participation of women in the workforce.

    1. For length, I had to cut the bit quoted below, and on second thought, I'm including it, as I'm a big believer that the Fed had to get control over the excesses in the financial sector in that 2004-2006 period, but I will always wonder if there wasn't a way they could have done so without crashing the whole system. I find this kind of counterfactual particularly difficult because I find it hard to imagine the levers that were at the Fed's disposal at the time, that is, prior to all of the additional regulation and knowledge that was only put in place and gained after going through the GFC.

      "And while certain factors were particular to this bubble, e.g., of fraud and households exceeding their borrowing capacity and ability to pay, we will never know to what extent a more optimal monetary policy and credit growth might have limited the harm done by those factors (I suppose the cartoon version would be “stop the bad stuff going on” and “slowly grow out of the problem”)."

    2. Thanks for reading! I will answer your questions (the best I can) in multiple comments.

      First, regarding the demographic cause of inflation, there is much more than just the evidence from the US. Countries that had larger increases in the number of women had higher inflation, and those that had smaller increases (e.g. Germany) had lower inflation. But additionally, there is an example of a country that has had more than one demographic transition ... Japan:

      Japan's demographic transition paused in the 1990s, but started again (along with inflation) in the 2010s.

      Now there is the narrative that inflation caused women to enter the workforce (incomes needed to increase faster than inflation, so women started working), but in every case the demographic transition precedes the inflation.

    3. Regarding the Fed raising rates from 2004-2006, I agree it is plausible that could be a causal factor. Raising interest rates could have made mortgages more expensive, eventually bursting the bubble.

      The problem with this is the timing of the shocks to construction hires versus job openings. If it had been job openings first, then the monetary policy explanation would have made a lot of sense. Employers could foresee a contraction (or more difficulty/expense setting up the credit to buy/build houses), and reduce the number of job openings. However, job openings in construction don't start declining until much later indicating employers saw the market as at least normal, if not expanding (there's a surge in construction job openings in the years between the decline of construction hires and the bubble popping).

      The fact that construction hires leads the recession while openings are increasing makes any kind of tight money argument much less plausible, and points to a scarcity of workers rather than a lack of demand for construction.

    4. In general, regarding credit, credit liability level continued to increase for households and businesses until 2008. This is after the Fed had paused for over a year and even after the Fed started to cut rates. If the increased rates starting in 2004 reduced the availability credit, it doesn't happen until 4 years later. No Fed models (and no macro models in general) can predict that far in the future, so there's no evidence there was any method here — nor evidence for any post hoc explanation in terms of models of credit and interest rates.

      It is possible that the Fed might have been able to pop the housing bubble if it had raised interest rates much higher, much faster. The rates of the 1980s did seem to have an effect on housing (my own parents didn't buy a house until mortgage rates came back down in the mid-80s). But as it is, Occam's razor suggests the null hypothesis should be no impact from monetary policy.

    5. Thanks. The advantages of having done the research are clear. I think I will have to do some work myself to attain similar levels of confidence (something like how it's always easier to believe something when you've finally seen it with your own eyes).

      As for the level of credit rising into 2008, it is a surprising phenomenon, but I recall when I tried to look only at mortgages, and then tried to adjust for the rise in housing prices per Case-Shiller, the result was that the residual growth (something akin to "volume growth") fell dramatically in the 2004-2006 period. I don't mention this because I can make a compelling case, but to suggest maybe one last box to tick if you ever return to the question and want to satisfy yourself that credit didn't lead.

      Of course, leaving aside Openings for a second and focusing on Hires, that they would flatten out first makes sense even if it would only mean being consistent with higher rates (which only got back to pre-2000s levels in 2006) and expectations of lower starts (which, for one-family units, appear to have had a dip into negative y-o-y as early as 4Q04 before some lower growth until moving to double-digit declines in 2Q06 and beyond. Of course, you yourself flag Starts as the factor that leads all but Hires, so I'm still probably only debating the point that rates, credit, and expectations concerning both may have caused the declines in Hires and Starts.

      Finally, to return to Openings, my only question is whether you don't think it dangerous to read too much into them. I don't understand why it should be a series subject to any kind of economic discipline, unless i misunderstand the costs associated with posting or holding open an Opening according to the way the statistics are measured. But if your research has demonstrated the informational content, validity, and value of the series, I won't be one to argue. Instead, if I have to accept that Openings are informative, I would suggest looking at how nonresidential construction spending was what was rising into the recession, while residential construction spending appears to have peaked in 1Q06. The same phenomenon should cause one to look back at construction Hires, as well, for if total construction Hires are falling while nonresidential construction is robust, it might mean that residential construction Hires must have been collapsing over the period. Unless you have a way of separating out Openings and Hires (and perhaps, similarly, credit) related to residential construction activity from those related to nonresidential construction activity (and in the case of credit, all other investment activity), such considerations would seem to complicate the story you've presented for the 2004-2008 period, I think.

      I would say, at this point, I am more persuaded of the connection that your work seems to have uncovered between the labor force and inflation – and I still think it's one of the coolest things to have come out of your research. I wonder, did my thought experiment of nominal income preceding productivity and real economic output, thus leading to inflation, in the event of sharp increases in the labor force seem plausible to you or was it just another one of those "economic cartoons" that ended up in my head? Not that an explanation needs to be forthcoming or is important – I have gained a greater appreciation for the concept of "effective theory" since reading your work and following where the ideas appear to lead. Thanks, again!

  2. I don't have the mind to hang with the math, physics, and engineering you link to (Terry Tao on advances in compressed sensing research!), but I have been skimming some of the economics links in the hope of making my questions a little more intelligible.

    Another question that occurred to me recently is whether you identify at all with some of the principles that purport to underpin real business-cycle theory. I say purport, because you might disagree with most of the theory and its predictions as understood conventionally, but with your dynamic information equilibrium models doing the work of explaining how an economy works normally, is it accurate to say that, like RBC, you explain nearly all deviations from the norm by having recourse to real shocks in your models?

    Although, perhaps a further point of departure from RBC would be in the characterization of the shocks. My sense is that all of the shocks in your model are "real" shocks, but in the information equilibrium paradigm, some meaningful proportion of the shocks will be the result of purely random processes, where correlations go up, entropy down, and economic activity excessively expands or contracts (but typically contracts, in light of less economically dynamic and robust conditions). Is that a sensible characterization of your theory and models, in which some of the real shocks will be random and some will be "non-random" (e.g., identifiable labor force shocks, identifiable energy/commodity shocks, etc.)?

    Finally, would you say that a DIEM perspective could be consistent with a view that markets don't exactly "fail" in the sense that they could be made "better" via effective policies? That seems to be another tenet of a certain kind of market fundamentalism, but I wonder if it doesn't apply to some extent to the picture of economics that your research presents.

    My motivation, if it isn't clear, is to feel I have the right context in mind when I try to situate and understand your work (and I do get that RBC might be relatively fringe relative to your engagement with the likes of DSGE models). Perhaps you could direct me to a post you've written in the past that helps with that, and perhaps you will temporarily write posts in the future that present the bigger vision – what your work purports to do, what it does not purport to do, what is the ideal objective that it aims at, and what some reasonable hopes and expectations you have as your research program continues to mature. I'm sure you are busy, but any further thoughts that you're able and willing to share would be greatly appreciated. Thanks!

    1. Sorry I missed this comment for some reason when it was posted.

      Regarding your question about RBC: RBC essentially calls the shocks "technology shocks". If you were trying to draw as close of an analogy as possible, the DIEM approach sees deviations from the norm as noise plus non-equilibrium shocks where we are agnostic about the cause. Only when you see similar patterns in shocks from different data sets can you start to see some evidence for a particular cause dependent on the timing. Often, it seems to be social factors. But political changes have some effects as well.

      Women entering the workforce seems to be behind the inflation shock of the mid 70s. The Affordable Care Act seems to have been behind a mini-boom in 2014. Australia seems to have had a commodity "boom" (a positive shock) around the time of the 2008 recession. The "Lawson boom" in the UK seems to have been a real thing (a positive shock).

      The shocks mostly seem to have real causes, but some are random in the sense that it's almost impossible to tell when a network of humans will go from optimism to pessimism (a "network cascade"). I like to say that humans are so complex that to a good approximation they are random (the definition of algorithmic complexity). However, sometimes they spontaneously coordinate in ways that affect the economy. Imagine most of the time, people are milling about in some plaza. Their distribution is going to be somewhat random from a high-level view (it's actually going to be complex, depending on how humans make paths through crowds and nucleate around, say, a cafe). But sometimes there's an event -- like say a famous person walking through the area that will suddenly coordinate the plaza (everyone starts looking over at the scrum of people).

    2. As for a bigger picture of my approach there's the free version of my book (i.e. the collection of blog posts that served as the first draft):

    3. Thanks for the reply regarding your concept of "shocks" and for the link to your book. The latter will offer a good opportunity to review your early posts. The former – the role of randomness in your theory being one of its greatest merits, in my modest opinion – creates an interesting tension between the implicit assumption in much of your work that a shock will eventually occur, be random, and presumably difficult to predict, and the work of yours that I have seen that might attempt to make some predictions of when certain kinds of shocks are will either be forced or at least relatively foreseeable (e.g., wage growth rates creating pressure on capital when they begin to exceed nominal output growth, or trying to see if a large enough deviation from trend in specific series are increasingly likely to be informative or even self-fulfilling, as seen in deviations in JOLTS data or the level of the S&P 500). Anyway, thank you for all of your effort and insights.

      On a specific point you reference – the 2014 mini-boom / wage growth shock – I can't resist mentioning that it is another example of the pattern I've been trying to highlight in my inarticulate way, which is an effect that you have seen in your work but that would appear to be "overdetermined" to an uninformed observer. I suppose I am only suggesting that it makes your job as a theorist more difficult and even annoying, as moving the priors of uninformed readers will be difficult and time- and energy-consuming. In this case, I am the uninformed reader, curious to know how or why you've settled on the ACA explanation (e.g., which data and how significant) and how you rejected or found negligible "contemporaneous events" (in this case, I'd just be playing devil's advocate, and might throw out, e.g., the shale boom in the domestic energy sector or perceived, although perhaps unfounded, "booms" in other "hot" sectors of the economy, such as heavy investment in Internet or life sciences businesses, again in the 2013-14 time frame). I assume you can hear the echoes of my questions concerning credit growth vs. demographics in both 1970s inflation and 2000s housing, which roughly took the form of "how did you size and time the impacts of factors X and Y such that you determined factor X deserves credit as the 'explanatory' factor?"

      Anyway, I feel I've already taken too much of your time, so no further response is necessary, with your responses thus far being much appreciated. Separately, I look forward to a deeper and presumably much longer discussion of some of those social and political factors that contribute to some of the shocks in our economy to which you merely allude, up to and including perhaps the biggest factors, having to do with the relative power of capital, labor, government, etc., and what, if anything, can be usefully said or done about them.


Comments are welcome. Please see the Moderation and comment policy.

Also, try to avoid the use of dollar signs as they interfere with my setup of mathjax. I left it set up that way because I think this is funny for an economics blog. You can use € or £ instead.

Note: Only a member of this blog may post a comment.