Thursday, May 8, 2014

Modern macro, without the math



Review: Kartik B. Athreya, Big Ideas in Macroeconomics: A Nontechnical View

Athreya is an economist at the Richmond Fed. The primary contribution of the book is to describe the workhorse analytical framework of mainstream macroeconomics without the use of any technical apparatus (math). This book does not have a single equation. Two types of people can reap large benefits from Athreya's efforts: (1) people considering or already pursuing graduate school in macroeconomics, and (2) outsiders to economics proper who want to read and respond to academic work in macroeconomics, like journalists or policy observers.

When I was studying for comprehensive exams I read Hal Varian's undergrad micro text. I had never really learned micro without the calculus (my undergrad econ program really liked math). Varian helped me develop intuition that gave me a better grasp of the math (which gave me better intuition). I think Athreya can serve a similar purpose for macro (and even parts of micro). I find math really useful, but the critics are right that it can sometimes cause us to miss the forest for the trees. Athreya can help you keep the big picture in mind by putting most macro models into the perspective of a single framework, and it can help you get more intuition out of the math itself. I will recommend it to undergrads preparing for grad school and those cramming for comps.

The second audience I mention--journalists and other observers who like to read and comment on economic research--are sometimes inclined to be critical of how macroeconomists do things. In some cases this may be driven by honest ignorance; modern macro isn't always very accessible. Athreya eliminates that problem. The book isn't an easy, casual read by any means, but anyone who wants to learn the material can do it. Athreya provides a lot of perspective on the tradeoffs faced by macroeconomists, including popular targets of ridicule like rational expectations, DSGE, and microfoundations generally. It would be refreshing if critics had a better grasp of the methods they criticize (though a few do); the tradeoffs faced by practitioners are steep. Athreya won't convince everyone, but at the very least I think it will dissuade any honest observer from acting as if macroeconomists are stupid and stubbornly obtuse. Macro analysis is harder than it looks (or, it only looks easy to the absurdly overconfident).

You could build a serious independent study of macroeconomics on this book and the many references Athreya suggests.

***

Athreya describes modern macro by building from the foundations of preferences and trading arrangements. He focuses on Walrasian equilibrium and the Arrow-Debreu (and later Radner) models. This proceeds to discussion of Pareto efficiency and the Fundamental Welfare Theorems. Along the way, he provides literature and insight into whether key assumptions are useful. He defends the use of math, which clarifies causal linkages and makes it so that our arguments are "over the relevance of preconditions, not over the conclusions given those preconditions" (195). As I've said before, we're all using models; some of us use transparent ones.

Athreya makes it clear that the caricature of macroeconomists as worshipers of Laissez-faire is misguided, but he also provides intuition for why many economists respect decentralized markets along many dimensions:

The "market" is acting as if it has solved a system of nonlinear equations with as many equations and unknowns as there are consumers and producers (as well as one more equation for prices). . . . These equations arise from the optimization problems of consumers and producers who take prices as given but know nothing about any more than prices and the equations describing what is optimal for them alone! (91)

This is basically the I, Pencil insight, or the insight we get from that guy who tried to build a toaster. It's very profound.

Athreya outlines conditions under which the models tell us to worry about the efficiency of decentralized outcomes: intertemporal issues (what he often calls IOU markets), monopoly, irrationality, public goods, insurance, and externalities. While things like rationality and monopoly attract a lot of attention, Athreya makes a compelling argument that these issues are far less problematic than market incompleteness (70). He argues that assumptions like rationality are good for protecting the public from economists, as departure from such assumptions can make room for a lot of arbitrariness. He also outlines the methodological tradeoffs associated with various modeling assumptions; these should be carefully considered by the critics (185). More broadly, he tries to give readers a feel for a central tradeoff in the study of macroeconomics, which requires researchers to reach "a compromise. . . between expanding the 'reach' of a model and retaining its internal consistency" (159).

Much of the book, like much of modern macroeconomics, highlights market failure. "A huge amount of work in modern macroeconomics is about studying the size of departures from efficiency" (334). Athreya focuses on search models, incomplete market models, and overlapping generations models as the main classes of models that fail to deliver optimal allocations from decentralized trading (276). Note that the topic of economic inequality fits well within the framework of some of these models, particularly the incomplete markets model, and Athreya discusses some of the relevant literature and the broad policy ideas implied by the main models. Incomplete markets models typically require some heterogeneity (a reminder that representative agent isn't the only game in town), and Athreya discusses how that can be modeled and what it can teach us.

Athreya also briefly describes the debate about calibration and estimation (263). "Calibration is thus not new, nor is it the preserve of one 'kind' of macroeconomist, nor is it politically stilted" (265). I could have used even more discussion of this debate. I don't have a dog in that fight, but in my main work I calibrate for computational feasibility reasons.

The book ends with a discussion of the Great Recession. By this point in the book, readers will already have an idea for how the standard tools can be used to study the crisis. Athreya gets more explicit, breaking the study of the crisis into three key questions: (1) Why did asset prices rise so much? (2) Why did initial changes get amplified? and (3) Why has the recovery been so slow? Broadly speaking, modern macro fared better than its critics allow when approaching these questions. However, Athreya does note some issues on which it failed (358): We know far too little about the rich structure of credit markets and instruments, the role of underwriting standards, the magnitude of the collapse of asset prices, the role of mass mortgage renegotiation, and others. Athreya concludes that we have plenty of work to do, but "these failures do not make a wholesale revamping of macroeconomics a bright idea" (359). Economists have plenty of incentives to develop empirically disciplined, persuasive responses to recent events, and this can be accommodated by the dominant framework. If you are dissatisfied with the macro field, Athreya may ask you to first consider how the changes you would like to see can be implemented on the existing, flexible framework. Note, though, that one need not agree with these conclusions to enjoy the book.

Being partial to firm dynamics, I would have liked to see more about how heterogeneous firm models can help us think about growth and business cycles. Entrepreneurship and firm dynamics generally may have a lot to say about inequality, the response of the job market to business cycle shocks, and the forces of creative destruction that we see clearly in the data but not in most macro models. The composition of the firm distribution is changing over time, and we need to know what that means and whether we should worry. The book is full enough already, so I don't blame Athreya for leaving this out, but it is my hobby so I had to mention it. I also think I could add a more specific item to the "how macro failed" list. Matteo Iacoviello describes having a paper rejected in 2001 with the following explanation:

This paper focuses on a small niche--the housing market--with limited evidence that this market has the significance that is implied for real economic activity.

I think we missed the boat on that one, though there were certainly a few people writing down models with housing, land, or durable goods. An asset with its own investment category and unique collateral and consumption roles deserves attention by macroeconomics. It's getting that attention now, and the methodological framework Athreya describes is perfectly capable of handling it.

***

Athreya has written a useful book. There was a need for a nontechnical explanation of the macroeconomics discipline. Even people with undergrad economics degrees sometimes don't know what macroeconomists do. It's too easy for outsiders to dismiss these tools without understanding why we use them. And for insiders, this is an excellent way to build intuition and refresh memory. You can think of it as a resource for grad students, or you can think of it as a thorough, constructive response to macro's critics. In any case, it's a nice addition to a bookshelf.

4 comments:

  1. Great review. Thanks for bringing this book to my attention. Understanding macro arguments has always been a relative weakness for me. Gonna buy it ASAP.

    ReplyDelete
  2. Nice review, Ryan. I had been meaning to order the book and now I have. I would add one more group to your list ... people who do research or policy work using these kinds of models. Thinking in 'general equilibrium' is not intuitive for most and yet that is the value-added of macro. If there's a book that shows economists how to convey the intuition of these macro models well without equations and tons of jargon, that's a real addition to our communication toolkit. I would argue economists are less likely to 'miss the boat' when others have a better idea of what boat we are on in the first place.

    ReplyDelete
    Replies
    1. Claudia! Yes, I hope the book would be helpful to macro policy people. At least it can help everyone speak the same language.

      Nice to hear from you.

      Delete
  3. Thanks much Ryan. This was a thorough and clear review. On specifics, it’s hard to disagree that firm dynamics (starting with Hopenhayn etc) might have deserved more space. One possible route would have been to lower the detail on the preparatory microeconomics, and add more on firm dynamics. I erred on the side of the former since I anticipated that many lay readers (journalists especially) would've wanted more justification for the type of analysis we do, but I can see it not being the only call.

    Re: housing—indeed. I liked that Iacoviello made the point he did—and our understanding of housing—and especially its relationship to other credit markets, was just not deep enough. But it is getting deeper fast (judging by the huge number of papers I have seen presented in the past four years). I also think it’s important to appreciate the progress that has occurred. After all, debt, especially of the short-term variety is a (the?) key, and as I stress in the book, until Townsend (1979) we lacked a theory for it at all. This is one instance that I think illustrates how deeply problematic are claims that microeconomics is some separate thing, and that "recession-macro" is somehow a finished intellectual project.

    I’d add also that one sub-text (since I don’t make it explicit all over the place) I hope readers are struck by is the speed with which the profession reacts to its theories’ collisions with data. Between the equity-premium and Shimer puzzles alone, it is obvious the extent to which macroeconomists seek models consistent with data, and that much of what we do is driven by an empirical challenge. And the competition to be first in these races is, as you hint (correctly, in my view), intense.

    I’d also be personally delighted if readers come away with was a sense of the extent to which we've been grappling with inequality and heterogeneity. Too bad we didn't tell anyone about it :) A nice thing is that some of that work (Victor’s and Tony’s) is surfacing (a decade or more late) to the broader public because of the buzz around Piketty’s book.

    Again, thanks for reading the book carefully.

    Kartik

    ReplyDelete