Bookblogging:Micro-based macro-introduction

by John Q on September 17, 2009

I’m starting now on what will I think be the hardest and most controversial chapter of my book – the argument that the search for a macroeconomic theory founded on (roughly) neoclassical micro, which has been the main direction of macro research for 40 years or so, was a wrong turning, forcing us to retrace our steps and look for another route. As always, comments and criticisms accepted with gratitude.

Refuted doctrines

Micro-based macro

We are now all Friedmanites, Lawrence Summers, (former US Treasury Secretary, now Director of the White House’s National Economic Council, and prominent New Keynesian economist) and

At the end of the 19th century, British Liberal politician Sir William Harcourt observed “we are all socialists now”. Harcourt was referring to a radical land reform measure that had been denounced as socialist when it was introduced, but was generally accepted by the time he was speaking (a couple of years later). Harcourt’s point was applicable to the whole trend of economic and social policy, in Britain and elsewhere, from the 1867 Reform Bill that gave millions of working class men the vote (women had to wait until after World War I) to the crisis of the 1970s.  From progressive income taxes to publicly-owned infrastructure services (both prominent items in the 10-point program of the Communist Manifesto)) ideas that were unthinkable in mainstream politics became issues of political contention and then established institutions.

 As a pithy summary of the way ideas that were once radical become acceptable, and are ultimately embodied in conventional wisdom, Harcourt’s quip has never been bettered. As a result, it has been reused many times over.

One of the most notable adaptations of Harcourt was that of Time Magazine in 1965, which noted, following the successful use of fiscal policy to stabilize the economy that “we are all Keynesians” now. This statement was made by Keynes’ greatest modern critic, Milton Friedman (though he later said it had been taken out of context). Even more famously, it was repeated by Richard Nixon in 1971.

But whereas Harcourt was speaking at the beginning of a nearly a century of reform that did indeed take economic policy in a socialist, or at least social-democratic direction, Nixon’s statement marked the end of the era of Keynesian dominance. 

In fact, Nixon was citing Keynes’ aversion to the gold standard (a “barbarous relic”) as a justification for abandoning the pegging of the US dollar to gold, which was a central feature of the Bretton Woods system of fixed exchange rates that had underpinned Keynesian economic management since World War II. The outcome was not a system of stable exchange rates backed by a basket of commodities rather than gold, as Keynes had proposed, but the complete breakdown of Bretton Woods and a shift to the floating exchange rate system advocated by the greatest critic of Keynesian economics, Milton Friedman.

In the course of the 1970s, Friedman and his supporters, centred on the University of Chicago, won a series of political and intellectual victories over the Keynesians. Following the failure of attempts to stabilise the economy using Keynesian fiscal policy, governments around the world switched to Friedman’s preferred remedies based on controlling the growth of the monetary supply. Even though this did not work particularly well, and was later replaced by policies based on managing interest rates, the resurgence of the Chicago School was not reversed.  Their case against government intervention, both to stabilise the macroeconomy and to address market failures in particular industries, was widely accepted.

The Keynesians conceded Friedman’s central points: and that macroeconomic policy can affect real variables, like the levels of employment and unemployment, only in the short run. They sought to develop a ‘New Keynesian’ economics, by showing that, given small deviations from the competitive market assumptions of the basic neoclassical economics model, it would be possible to explain the recurrence of booms and recessions and to justify the modest stabilisation policies pursued by central banks during the Great Moderation. Because prominent representatives of this group were located at Princeton and Harvard on the East Coast of the US, and at Berkeley on the West Coast, they were sometimes called the ‘saltwater school’ as opposed to the ‘freshwater school’, located in the lakeside settings of Chicago and Minnesota.

Members of the freshwater school sought to push Friedman’s conclusions even further, arguing that macroeconomic policy could not be beneficial even in the short run. They tried to show that government intervention could only add uncertainty and instability to the economic system, and that, in the absence of such intervention, economic fluctuations like booms and slumps were actually good things, reflecting economic adjustments to changes in technology and consumer tastes. The resulting models went by various names, but the most popular was ‘Real Business Cycle Theory’.

Despite their often heated disagreements, saltwater and freshwater economists were in agreement on one fundamental point: that macroeconomic analysis must be based on the foundations of neoclassical microeconomics. And, although they disagreed about economic policy, these disagreements could be contained within a very narrow compass. With a handful of exceptions, both schools took it for granted that macroeconomic management should be implemented through the monetary policies of central banks, that the only important instrument of monetary policy was the setting of short-term interest rates and that the central goal of monetary policy should be the maintenance of low and stable inflation. Granting these premises, saltwater economists argued that stability could only be achieved if central banks paid attention to output and employment as well as inflation, while the freshwater school favored an exclusive focus on price stability.

The global financial crisis did not so much confirm or refute the elaborate arguments of the competing schools as render them irrelevant. The saltwater school could claim vindication for their view that the economy is not inherently stable, but their models had little to say about the kind of crisis we have actually observed, driven by an interaction between macroeconomic imbalances and massive financial speculation.

The freshwater-saltwater disputes were similarly irrelevant to the policy debate which was conducted in terms that would be familiar to someone who had not looked at an economics book since 1970. (In fact, the freshwater side of the dispute rapidly reverted to arguments from the 19th century, which had been debunked by Keynes and Irving Fisher).

 As Gregory Clark of UC Davis observed ‘    The debate about the bank bailout, and the stimulus package, has all revolved around issues that are entirely at the level of Econ 1. What is the multiplier from government spending? Does government spending crowd out private spending? How quickly can you increase government spending? If you got a A in college in Econ 1 you are an expert in this debate: fully an equal of Summers and Geithner.’

If we are to develop a macroeconomic theory that can help us to understand, and hopefully prevent the recurrence of, crises like the current one, and help us to improve policy responses, economics must take a different road from that it has followed since the 1970s. The appealing idea that macroeconomics should develop naturally from standard microeconomic foundations must be recognised as a distraction. In its place, we must accept, in the language of systems theory that macroeconomic phenomena are emergent, arising from complex interactions of behaviors we do not fully understand, but must nevertheless respond to.



Thorfinn 09.17.09 at 4:24 am

As Friedman might say; it shouldn’t matter what the foundations of your model are, as long as it explains the data.


Substance McGravitas 09.17.09 at 4:27 am

We are now all Friedmanites, Lawrence Summers, (former US Treasury Secretary, now Director of the White House’s National Economic Council, and prominent New Keynesian economist) and

And what?


Substance McGravitas 09.17.09 at 4:34 am

They tried to show that government intervention could only add uncertainty and instability to the economic system, and that, in the absence of such intervention, economic fluctuations like booms and slumps were actually good things, reflecting economic adjustments to changes in technology and consumer tastes.

I’d like to see an expansion of this. If the point of the book is that the zombie ideas are bad for people then this here seems like a pretty critical deviation from what good and bad might mean to the folks I see walking up and down my street.


John Quiggin 09.17.09 at 5:34 am

@3 There will be more on RBC to come.


James Kroeger 09.17.09 at 9:29 am

If we are to develop a macroeconomic theory that can help us…to improve policy responses, economics must take a different road from that it has followed since the 1970s. The appealing idea that macroeconomics should develop naturally from standard microeconomic foundations must be recognized as a distraction.

I’m not sure where you want to go with this, but I would suggest that you focus on the assumptions that neoclassical economists make about inflation, which is the one economic aggregate that they believe the government should try to regulate, because individual economic agents are expected to respond to ‘the rate of inflation’ in a way that can affect the macro economy in a negative way.

They embrace the simplistic assumption that ‘the rate of inflation’ is something that affects all economic agents in the same way, perhaps because it has been common practice for economists to measure price inflation as an ‘average.’ If economists could be bothered to disaggregate the inflation rates they compile, they would discover that the rate of price inflation varies across different income ‘brackets.’ Indeed, it is entirely possible for the members of one income bracket (e.g., the wealthy) to experience a very robust rate of price inflation at the same time other income groups (e.g., the working poor) are experiencing disinflation or even deflation.

That is precisely what happened over the past quarter century in America, ever since Ronald Reagan decided to throw billions of disposable dollars at rich people at the same time that outsourcing/globalization/immigration put downward pressure on the wages of the poor. The rich did what you would expect with the extra dollars, which was to throw them at luxury and asset markets, setting off a round of price inflation for the people living in that income group, and that helped to create the instability that we saw in the asset markets.

Neoclassical economists have preferred to characterize this surge in asset price inflation as a positive economic benefit, implicitly suggesting that asset price increases in secondary markets somehow provide extra funds to firms for economic investments, but they are nothing of the kind. (They also conveniently ignore the empirical fact that roughly 85% of all investment by corporations is funded by retained earnings or other internal sources.)

Yes, they embrace a number of flawed assumptions that misrepresent the facts, but this one in particular explains why their models did not identify the kinds of disequilibria that ultimately contribute to systemic collapse in the financial sector. Good luck with your efforts, John Quiggin.


Tim Worstall 09.17.09 at 9:55 am

“The appealing idea that macroeconomics should develop naturally from standard microeconomic foundations must be recognised as a distraction. In its place, we must accept, in the language of systems theory that macroeconomic phenomena are emergent, arising from complex interactions of behaviors we do not fully understand, but must nevertheless respond to.”

I assume we’re saying that neo-classical micro is that standard micro and that it is “right” in the sense of correct.

Which leads to something of a puzzler. For if it is correct as a way of describing the incentives that people face and how they react to them then we would very much indeed like to base our macro, how people in aggregate react to incentives, on the same correct revelations of how humans react to the proddings of the universe.

To say “it works in micro but not in macro” is (to be very unkind about it) a little like a Creationist’s argument that sure, there’s micro-evolution but not macro-evolution.

If we’re saying that we don’t know enough about how it scales up from micro to macro so we’re going to have to use different methods and assumptions, then perhaps fair enough. But that would seem to argue for more research in how it does scale up rather than abandoning all hope.

If on the other hand we’re saying that we can never know how it scales up (say, because the system is chaotic or something) and thus can never relate micro to macro then doesn’t that leave us with certain problems of what to actually base macro on?


Arin D 09.17.09 at 9:59 am

I found myself agreeing with much of the article. And then I got to the prescription:

“The appealing idea that macroeconomics should develop naturally from standard microeconomic foundations must be recognised as a distraction. In its place, we must accept, in the language of systems theory that macroeconomic phenomena are emergent, arising from complex interactions of behaviors we do not fully understand, but must nevertheless respond to.”

This is both ” too facile” of a conclusion and “too burdensome.” Too facile in the sense that it sounds good, but it’s not clear that getting rid of microeconomic foundations must be the central desideratum going forward. Particularly because “micro foundations” can mean very different things to different people. In the late 60’s and 70’s, Clower and Leijonhufvud provided “micro foundations’ by using Walrasian general equilibrium theory but dispensing with the assumption (yes assumption) that markets clear everywhere and always. This stands in stark contrast to the rational expectations a la Lucas and future Dynamic Stochastic General Equilibrium modelers, who assume that to be a defining characteristic of “micro foundations.” This is one example, but I think it shows one way in which the statement above about needing to dispense with microfoundations is too facile.

And now we get to why it may be “too burdensome.” Because what is postulated as alternatives (“systems theory” where macro phenomenon are “emergent”) is not even clear enough to merit the role of the main alternative to existing macro. Again, this strikes to me as a “pet theory” phenomenon. When mainstream approaches seem to fail, everyone has their “pet theory” that must be the correct approach. For what it’s worth, several decades of researchers have tried to model macroeconomic phenomenon as “emergent” – in the sense of using agent-based modeling and elements of complexity theory. This has happened especially outside of mainstream macro – at places like Santa Fe Institute. But the verdict is far from clear that these approaches provide the clear way forward. Let me re-iterate that I don’t think that they are not useful. However, the prescription in your post needs more careful evaluation of these approaches (their successes and failures) and not simply a declaration of the obviousness of what we ” must” do.


Stuart 09.17.09 at 10:33 am

To say “it works in micro but not in macro” is (to be very unkind about it) a little like a Creationist’s argument that sure, there’s micro-evolution but not macro-evolution.

Couldn’t another analogy being that classical physics doesn’t scale down to the quantum level (and vice versa)?


Kenny Easwaran 09.17.09 at 11:33 am

A closer analogy would be to say that we can’t predict the classical scale phenomena given the quantum descriptions – which of course is completely wrong. Of course, a more congenial analogy is that even though we know that humans are physical systems, we also admit that there’s important room for an autonomous science of human physiology, that doesn’t try to predict everything about the behavior of the human body from the laws of physics that govern the fundamental particles making up a human. (Maybe some day we’d like to get to that point, and we can understand some specific features in those terms, but no one thinks we’re there yet.)

On a separate point, why is it important to Keynes to have the currency pegged to some basket of commodities? I can see why it’s important to monetarists to let the currency float free (because otherwise you can’t really effectively control the size of the money supply) but why would Keynesians be particularly attached to the Bretton Woods scheme?


Barry 09.17.09 at 1:28 pm

Thorfinn 09.17.09 at 4:24 am

“As Friedman might say; it shouldn’t matter what the foundations of your model are, as long as it explains the data.”

Except that an important goal of models (and theory) is to make good predictions, for new situations.


Joshua W. Burton 09.17.09 at 1:32 pm

Couldn’t another analogy being that classical physics doesn’t scale down to the quantum level (and vice versa)?

Yes, but this is nonsense of course. If classical physics didn’t emerge cleanly as the limit of QM when hbar goes to zero, nobody would be interested in QM. There are emergent phenomena in classical physics that are hard to explain quantitatively from the fundamental microphysics, such as for example the onset of turbulence, but while we can’t calculate everything we can calculate enough to demonstrate easily that the Navier-Stokes equations are at work (and that they need — indeed, will admit of — no large modification). In physics, the novelties and surprises go in the other direction: the classical limit of a fermion field is identically zero, for example, so while classical physics has photons (namely Maxwell’s equations, in the continuum limit), it has no electrons or protons, and therefore no explanation for matter that survives into the quantum realm. Similarly and still farther in, the Higgs mechanism offers an explanation for fermion mass that has no nonrelativistic limit, and so low-energy effective QFT has no explanation for its fundamental fields below the electroweak breaking scale. But the standard model has no trouble seeing downwards to ordinary particle physics; the vacuum is hard to calculate, but if the equations were wrong we’d notice it.

Physics envy in the social sciences is a little sad in my aesthetic opinion; I’m not qualified to have a technical opinion about whether it leads to productive lines of thought. But in this case, it can only lead economists astray . . . unless the reductionist goal of deriving an explicit limit (analogous to hbar -> 0) in which microeconomics becomes macro, is mathematically coherent, quantitatively testable and potentially achievable.


Yarrow 09.17.09 at 1:49 pm

The best capsule summary I’ve read of this crisis and of the Great Depression is from a historian: James Livingston said “The underlying cause of that economic disaster was a fundamental shift of income shares away from wages/consumption to corporate profits that produced a tidal wave of surplus capital that could not be profitably invested in goods production—and, in fact, was not invested in good production.” (There’s also a Part 2.) He cites increasing productivity of labor and decreasing power of trade unions as reasons for the shift in income shares.

This seems so plausible that I’m puzzled as to why it’s hardly mentioned, not even to be debunked. Because it suggests that the solution is to tax the rich? (Taxing the rich and giving the money to the rest of us would let the economy grow again. Taxing the rich and throwing the money in the ocean would at least prevent the kinds of bubbles we’ve seen recently and in the run-up to the Great Depression.)


SamChevre 09.17.09 at 2:04 pm

It seems that resource shocks are pretty key to both the 1970’s and the last 5 years. (Oil prices, in both cases.)


John Quiggin 09.17.09 at 2:15 pm

This is only the introduction. I plan to deliver a more detailed discussion of the way I think macro should go at the end of the chapter, which will hopefully be appearing here before too long.


Steve LaBonne 09.17.09 at 2:17 pm

Why is the foundation of neoclassical MICRO economics still regarded as sound? Since the first papers of Kahneman and Tversky (or since forever, if one has any common sense) it’s been untenable to hold that people are anything like consistently rational decision-makers in even the simplest contexts. I don’t know nearly enough to have any idea how far along the behavioral economists are at building a viable replacement theory, but for sure the psychological foundation of traditional microeconomics has to be regarded as pretty shaky. No matter how pretty the theories he makes possible, it’s an empirical fact that “homo economicus” just doesn’t exist.


wjd123 09.17.09 at 2:21 pm

I have a formula. There are as many types of human nature as there are societies. The more capitalistic the society the less social is economic man. Therefore in order to protect what is social from economic man in a capitalistic society, rules and regulations need to be inversely proportional to the type human nature capitalism creates.


John Cisternino 09.17.09 at 3:25 pm

I like this kernel of a chapter very much. I think you’re onto something significant in focusing on the micro-foundations move. The language of emergence that you allude to in the last paragraph is certainly evocative. It made me want to look back at Chomsky’s writings on reductionism in the sciences, which I’ve often found clarifying.

In the case of building micro foundations for macroeconomics, it seems to me that there are two enormous and related conceptual problems at the heart of this quixotic quest that should have rendered int a nonstarter and should make us feel good about moving on to other more productive pursuits.

First, macro wasn’t well enough worked out to know what exactly was phenomena needed to be explained through microfoundations. It’s hard to build foundations for our explanations and descriptions of macro-level phenomena before we have a good understanding of what those phenomena are.

Second, contra the implication of comment #6, classical microeconomics is based on axioms that everyone knows are false. Even Milton Friedman. The assumptions of of rationality, perfect information, etc. are simplifying assumptions. We know that real economic decisions are not accurately modeled this way (and there are many research programs that are engaged in building a better description of micro-level economic decision making). We justify our use of these transparently false assumptions by virtue of their usefulness (though no one thinks they yield good-enough-for-practical-reckoning predictions in the way that Newtonian Physics does), and by the lack of a worked out, better predictive alternative. But we don’t think they’re true. It seems daft to try to yoke macro to a set of assumptions we know are false.


Chris 09.17.09 at 3:45 pm

@15: Yeah, I wonder that too. It’s the equivalent of building your theory of hydrodynamics on the idea that water molecules are tiny billiard balls. Sure, it works in some ways as a first approximation. You can get some vaguely-correct idea of what pressure is and why water flows downhill. But water molecules *aren’t* tiny billiard balls and sooner or later, at some level of refinement, that is going to bite you in the ass.

This analogy also suggests that you don’t need to understand micro *at all* to start looking at macro on its own terms. The idea that water flows downhill, that there is such a thing as pressure, and how to make some kinds of pumps were understood before there was *any* substantial atomic/molecular theory of the microstructure of water, let alone an accurate one. Just look at the macro-level phenomena in terms of the properties of them that are measurable in their own right and see what you see.

A grand unified theory might be your ultimate goal, but there’s no reason it needs to be the immediate one. You don’t need to understand the molecular structure of water to build a levee or a ship, because those goals can be achieved even if your understanding of water is only at the macro level. (Osmosis is going to be mysterious until you have better microfoundations, but you can at least measure and describe it even if you can’t immediately explain it.)

In short, I think that while economists are prone to claiming that economics is a science, they are far less likely to take an actual scientific, data-first approach to practicing economics and most of them ought to be clouted about the head and shoulders with a copy of _Novum Organon_ until they understand why science is done the way it is done and not by starting with armchair theorizing. (I think this somewhat parallels Krugman’s critique of valuing theoretical elegance over fidelity to observable facts.)

Also, I agree with Yarrow that the shift in income shares between the working class and the investing class seems like a fairly large smoking gun in both the GD and current crisis and it’s odd that many economists apparently haven’t even noticed. Do they have a theory that proves that this can’t possibly be relevant, and if so, why aren’t they reexamining the theory in light of the facts?

It also seems to me that in light of financial innovations like credit unions and mutual funds, we don’t really need an investor class anymore; investible capital can do all the things it usefully does whether it is owned by a few people or by many people. (Of course there would still be a capital-managing profession, but in the absence of the current enormous corporate governance and principal-agent problems there is no reason they should make any more money than dentists or mechanics.)


salacious 09.17.09 at 4:20 pm


Isn’t the difficulty in deciding whether this shift in income shares is an exogenous cause or an endogenous consequence of processes driving the financial crisis? For example, if we want to go with the global savings glut account of the crisis, the increase in income for investors is obviously related to what happened, but it isn’t really causal in the way you’re implying.


stostosto 09.17.09 at 4:22 pm

We are all Quigginiks now.

One link between macro and micro is the velocity of money, reflected in the money multiplier. Friedman’s recommendations were underpinned by the observation that velocity had behaved remarkably stable over a long time span. Hence it made sense to perceive inflation as “always and everywhere” a monetary phenomenon. Say’s law, as I understand it, is assuming a strictly constant velocity.

Looking at the money multiplier :

It’s clear that things fell off a cliff last september. But they haven’t moved out of the doldrums yet.

What puzzles me in that graph is that the multiplier actually exhibited a long slide prior to its crash. I would have expected the opposite given the development of a huge shadow banking and derivatives market…

(Or maybe I have the interpretation of that multiplier wrong..?)


Barbar 09.17.09 at 4:43 pm

The more capitalistic the society the less social is economic man.

False dichotomy.


Substance McGravitas 09.17.09 at 5:05 pm

@3 There will be more on RBC to come.

It’s less “more on RBC” that I was concerned about, rather it’s the split I naïvely identify between policies more directly aimed at people (Communist Manifesto items noted above) and policies aimed at economic entities, the byproduct of which is hoped to be good for people.

Whatever I’m understanding or misunderstanding may come together with RBC, but in my ignorance I see in this passage a tether being cut and the balloon floating away.


Yarrow 09.17.09 at 5:21 pm

@Salacious: Isn’t the difficulty in deciding whether this shift in income shares is an exogenous cause or an endogenous consequence of processes driving the financial crisis?

To be sure; but my question is why the following plausible story isn’t really on the table, let alone the one to beat:
1) Productivity goes up.
2) The investing class (having the lion’s share of social power) captures the lion’s share of the excess. Profits go up.
3) The investing class needs to invest its new gains — and further wants to do so at the now-expected higher profit levels.
4) Production of goods can’t soak up the excess, because most folks have no more money than they did before, and therefore can’t buy more goods.
5) So there is enormous pressure to come up with paper profits, and more and more paper profits are essentially scams.
6) Finally the bubble pops.
Certainly there are many details to fill in; but seems plausible to me because it fits with the way capitalist society actually works.


salacious 09.17.09 at 5:53 pm


Some possible objections:

1.This explanation always works. Productivity rising is a consistent feature of modern economies. Your account doesn’t show why *this crisis* happened at this point in time, rather than at some other point. What changed to activate the series of events you describe?

2. The financial crisis occurred across a wide variety of countries, with a wide array of sociopolitical configurations. It happened in America and it happened in Europe too. Your second step seems like it would function differently depending on the particular balance of power present in any given country. In practice however, the financial crisis played out pretty much the same everywhere. (You might be able to rebut this by arguing to globalization/financial integration.)

3. There’s good reason to believe that the disproportionate profits accruing to the “investing class” was a consequence of the asset bubble you describe — with so much money sloshing around it was easier to skim off the top. Their insanely aggressive profit chasing may have helped sustain the bubble, but it’s not clear they started it. We might have to look elsewhere for the true cause.


Tim Worstall 09.17.09 at 6:06 pm

“The investing class (having the lion’s share of social power) captures the lion’s share of the excess. Profits go up.”

Well, Piketty and Saez (in their studies of income shares etc, top 10%, top 1% ) are pretty clear that in the 20s it was indeed capital income which fed the high incomes. In the past few years, it’s been labour income.

But perhaps it’s also possible that the “investor class” is in fact us, through houses, pensions, insurance policies and the rest.

Bill Gates has what, $50 billion? In a $15 trillion economy that’s 0.33% of GDP (yes, I know comparing a stock to a flow) something like 0.1% or less of US total wealth (maybe much less, I’m guessing). One of the Norman knights back in 1080 had 10% of England’s GDP as his private wealth (not a useful comparison, but just to give scale)


Akshay 09.17.09 at 6:16 pm

While we are discussing physics and reduction, let me recommend the article “More is different” by solid-state physicist and Nobel laureate Philip Anderson. It contains some very elegant formulations, such as:

“The ability to reduce everything to fundamental laws does not imply the ability to start from those laws and reconstruct the universe.”
“At each [hierarchical] stage entirely new laws, concepts and generalizations are necessary, requiring inspiration and creativity to just as great a degree as the previous one. Psychology is not applied biology, nor is biology applied chemistry”

PS: Of course, it’s worse in economics, as the microfoundations are untrue. Even if people “respond to incentives” (a tautology), this does not mean they are optimizing anything in particular.

PPS: Watch out with concepts like “complex systems science” and “emergence”. Even if some researchers have more-or-less precise definitions of these concepts, far more often they are used in a flaky manner. And anyway, you don’t need “complex systems science” (whatever that is) to argue against methodological reductionism.


Akshay 09.17.09 at 6:27 pm

Oh, and if you’re still writing the chapter on the future of macro, have you seen Dirk Bezemer’s article “No one saw this coming”? In the article he lists a group of economists and commentators who *did* see this coming and explains clearly what distinguishes them from the august institutions which did not.

To make a long story short: the Drs Panglos use general equilibrium models which are (duh) in equilibrium and contain no explicit model of the financial sector. The Cassandra’s use “old style” Keynesian flow-of-funds models/reasoning and have much interest in debt levels and the FIRE sectors.



sclimatus 09.17.09 at 7:12 pm

I think Keynesianism fell from grace because of 3 key criticisms: one academic, one political, and one empirical. Academically, the notion of “sticky prices” became trumped by rational expectations. Although economists may have overreacted in this direction, it was clear by the 60s that most prices were not so sticky as Keynes had argued. Politically, Keynesians had become dogmatic and closely tied to the financial institutions of the day, much as neoclassicists are now. With that dogmatism came a belief that a strict application of consensus economic approaches would soon solve all the problems of macro policy (sound familiar?). Monetarists were striking back at this approach and were often inspired by Austrian economists who believed that economic actors had more knowledge about their condition than the government did. Empirically, the Phillips Curve, a key instrument for explaining policy because it was so easy to understand, was entirely useless for understanding the “stagflation” of the 1970s. In fact, the whole Keynesian approach had difficulty explaining what was going on (here Gordon argues that the post-78 Keynesians represented the apex of saltwater macro). These issues, combined with political factors, led to a replacement of Keynesianism with Monetarism.


evil is evil 09.17.09 at 7:44 pm

I am sure that I have commented here before that any and, I mean any, economic theory fails when it cannot accurately measure the huge amount that it ignores in the blackmarket, the hidden taxes by the Mafia, the corruption of the permanently upper class in the nepotism and class ties, the ability of unregulated con artists like Bernie Madoff to steal and waste billions, the flight of money to “off shore” tax havens, etc, etc, too freaking etc.

There is not and never will be a free market.

This is not a democracy, it was set up solely for the benefit of the ruling class when the Constitution was accepted. This is an oligarchy and, by definitions that I learned when I studied economics, it is a monopoly market, where the prices are fixed by the largest corporations. More efficient and effective members can not enter.

A perfect example of this is the “defense” industries. I, once, bid on a tiny project for the military procurement administration. I won the bid and built what I said I would build. I did not get paid for at least four years. Rigged for the merchants of death and I was not a member of the “club.”

Also, no encompassing economic theory that I have ever read flat out says that the “defense” industry must be regarded as almost total waste and, only, shifts money to the rich. If the plutocrats were taken out of the equation completely and all “defense” industries were closed and the laid off workers were paid their full pay and allowances, we would get exactly the same result without giving the richest 1 per cent what is currently their share.

Finally, any economists that even thinks about the “trickle down effect” has never comprehended the history of mankind. The sole things that the rich want to acquire is land, servants and government bonds. None of these is anything like productive. All historical research show that the only thing the rich want to do is keep what they have stolen.

Also, no economic theory ever addresses the problem of propaganda. Any so called rational buyer would certainly never have participated in the farce in Washington, DC, where they were protesting against things that are not in their best interests.

This is due to the propaganda that they are provided by the “news” organizations that are wholly owned and controlled by the plutocrats.

I have no clue of why I even open this blog. Pure masochism, I’m sure. Possibly resentment because of the time in my finite life that I wasted spending money and part of my life studying economics.


Chris 09.17.09 at 8:10 pm

Well, Piketty and Saez (in their studies of income shares etc, top 10%, top 1% ) are pretty clear that in the 20s it was indeed capital income which fed the high incomes. In the past few years, it’s been labour income.

Labor income in the top 1%? Only if you count CEOs as “labor”, or maybe certain sports and entertainment stars. (It’s not entirely clear to me how you *should* count professional corporate managers, who receive part of the wealth produced by the corporation while often contributing dubious or negative value to it; they seem to be a kind of economic parasite on the corporate organism. But “labor” as the term is ordinarily used seems far from adequate.)

Salacious’s objections at 24 are interesting, but I think you could deal with 1 and 2 together by supposing that the U.S. was the 800-pound economic gorilla behind the GD and modern crisis, and the relative power of different economic classes in the U.S. varies over time; so when the U.S. wealthy capture a very large share of income gains, they produce an asset bubble and burst that afflicts everyone, and other times, they don’t. Other countries can and do have localized crises, but only the U.S. is big enough that when it falls, it causes an earthquake and makes everyone else stumble.

That, of course, depends on the idea that the U.S. really was an economic 800-pound gorilla basically from WWI on. I don’t know how accurate that is.

Objection 3 seems like a chicken and egg problem, but maybe a careful study of time series could resolve it. Did income inequality rise before or after, say, Case-Shiller or the stock bubble?


spencer 09.17.09 at 8:39 pm

I agree with Chris in that the old clear division between labor and capital income has not been that clear in recent years. Do we really want to counts the massive increases in income to the CEO class as an increase in returns to labor?


Adrian 09.17.09 at 9:01 pm

As I understand things, behavioural economics uses psychological results to improve the model of individual agent used in micro. How is that going to feed through to macro (and macro-based policy) unless macro looks to the newer, truer basis of micro as foundation?


Mattyoung 09.17.09 at 9:14 pm

“What is the multiplier from government spending?”

You lost me. Why is the government even mentioned in Econ 1? Econ 1 should be about stability in aggregate systems of individual agents. Government should only come into the picture in Econ 233 when the upper division students begin modeling fat tail characteristics.


John Cisternino 09.17.09 at 9:19 pm

@ 31 -I think we best feed micro insights into macro by viewing both research programs as equally based in observation of aspects of reality, then looking for opportunities to build conceptual bridges as we go. Behavioral economics hopefully will help us do better. But it’s still a long way from being a unified theory of decision-making. And even if it were, the scientific unification project will be intellectually much stronger if instead of looking at one body of knowledge as conceptually privileged over the other, we seek to integrate and correlate observations that have equal claim to legitimacy. That way we don’t deform our description of the macro-phenomena to fit a Procrustean bed. We lack a priori grounds for asserting that macro phenomena just are aggregations of microphenomena. By assuming so, we needlessly put roadblocks in the path of inquiry.


Robert 09.17.09 at 9:27 pm

Income inequality rose before the bubble. During the 1950-1970 period, we had wage compression, high productivity, and interest rates were below GDP growth rates, and the middle class (e.g. third quintile as measured by the census, or really all of the quintiles) grew with GDP. High top rates and big government spending helped.

In the 1970-1982 period, we started to have dispersion. Basically all cohorts except the top decile could not keep up with GDP, and median earnings for males became flat. They never exceed 1970s levels, in real terms. Household income grew somewhat as more women joined the workforce and their wages started to catch up with men, but in terms of labor input per adult, real median wages never exceeded early 1970 levels. There is some debate about counting benefits and whether you should discount with PCE or CPI-U, but I am referring to cash wages discounted by PCI-U, which is what you need to buy things with.

Take off occurred in 1982, with the start of a long process of swelling debt, falling interest rates, and growing inequality.

Stock Bubble — ending in 1987 crash
LBO bubble — Remember Gordon Gecko?
Fortunately, housing started booming in 1986, due to the mortgage interest deduction.
This ended in 1990 with a national housing recession and the banking crisis — our first post-war credit crunch. Remember RTC?
Housing bottomed in 1996 — fortunately, that was the year that Greenspan made his “Irrational Exuberance” speech about the stock market — Oh, forgot to mention that stocks took picked up the slack from housing.

During all this time, the top 1% shares kept growing, except for the credit crisis of 1990. By about 2000, the third quintile had lost half a GDP in wealth and assumed half a GDP in consumer debt since 1970.

During all this time, productivity growth was declining (except for “financial services”), GDP growth rates were less than interest rates, and both were falling. The bottom 90% of incomes were stagnating, and consumers were taking on debt to make up the difference.

Alas, it’s hard to repay debt in that case (you need wage growth rates to be higher than the rate of interest). Fortunately, rates kept falling, so the debt service ratios were not too high until the 2000s, and all the newly minted asset millionaires were happy to roll that debt over and keep spending humming.

Housing regained its abnormal rise, recovering from the 1996 lows in only a couple of years.
Stock bubble again with the .com blow out.

Recession, with a new housing bubble and here we are. A big Debt to GDP ratio, a big debt-service ratio, income inequality levels that exceed 1929, and rates are at zero.

It’s going to be hard to kick start those rates and climb out of this, because rolling over debt is a pain when rates climb. Also, asset prices tend to fall in those cases. Absolutely inequality increased before this crisis. The last 20 years — really the last 10 years have been about only one big thing — how to recycle earnings via debt in order to keep wages low. It’s the great recycling problem. Micro people will tell you that when wages fall, so do prices, and everything adjusts.

But we live in the era of debt-financing. The Chinese can recycle their surplus back to us by means of debt, and currencies don’t adjust. We can keep buying. Is it irrational for them to send us goods? Is it irrational for us to buy them? Maybe when the microfolk can define these terms better, we can decide whether there is a single “utility” function that a nation can be said to be maximizing, and how this function describes the 1982-2007 period. It’s all phlogiston to me.

In the same way, the top 1% can recycle their surplus back to the lower cohorts, who were eager to take on more debt. So, a way has been found to both lower wages for the majority of consumers and yet keep their spending high, at least for 20 years. And this combination of lowered wages and high spending was very good for profits, for assets, and for inequality. Dividends/corporate profits also reached 1929 levels, nearing 4%/6% of GDP.

We’ll have to wait and see how eager the poorly paid cohorts are to borrow, going forward, and what this means for prices.


Marc 09.17.09 at 9:33 pm

Should the goal of an economic theory be to prevent an economic crisis? I hope you meant to say that the goal of an economic theory should be to reveal if a certain economic development is based on sound or shallow fundamentals. Malinvestment leads to crises, inevitably. Not having the crisis when it’s due merely delays it and magnifies it.

So the goal should not be to introduce less uncertainty in our system, but MORE.


John Quiggin 09.17.09 at 10:50 pm

One point I plan to bring out is the behavioral deviations from rational optimisation that are important in micro terms are not the same as those important in macro terms. So, micro will use one set of simplifying assumptions and macro another.


Charles Peterson 09.18.09 at 3:36 am

Man is social. What do economic theories have to say about society? Society is composed of various groups of people. It seems that groups of people can accomplish far more than what separate individuals can. In real life, people are connected to many different groups in many different ways (such as family, extended family, friends, colleagues, co-workers, schools, churches, companies, unions, countries, alliances, etc.) Sometimes people receive great benefits from being part of groups, and other times they make great sacrifices for them. What do economic theories say about such groups? Do they predict benefits from group or collective actions?

From what little I understand, neither neoclassical nor Keynesian theories have much to say about groups. Failure to address the most fundamental aspect of human nature does not bode well for this enterprise.

And this thing that humans are often claimed to be by economists, “rational” (and individualistically selfish), seems to be more notably absent than present. People seem more likely to follow others in some way or another than anything else, and to promote the values they feel are best for everyone, or at least everyone in some group.


JoB 09.18.09 at 8:27 am

John, interesting post-wise but: will it come together as a book? Maybe you should share some kind of outline. I’m lost whether your point is beating dead horses, building a new race track or analyzing why the old race track led to so many injuries.


Adrian 09.18.09 at 10:29 am

@John Q: great, I look forward to learning about that. @John C: my thought is partly the quasi-ironic one that the appeal to micro-foundations was a freshwater thing not because there’s something freshwater about reductionism, but because micro was where the rational agent lived. If micro’s getting fixed, there’s no need to throw out the metaphysical baby with the Chicagoite bathwater. The demand that macro be reconcilable with micro, in the abstract, has merit. That said, your approach sounds good; there’s no reason to think macro has to do all the reconciling.


Chris 09.18.09 at 2:17 pm

@38: They should come together at some level eventually, but there’s no need to demand it up front. To take a different analogy from the physical sciences, chemistry as a science achieved many useful and interesting results even before subatomic physics had advanced enough to provide it with a decent understanding of exactly what a “covalent bond” was. The periodic table started out as an *empirical* collection of observations of the properties of various elements and how they lined up, and was connected to electron orbitals only after the fact.

Looking at macro on its own terms might tell us something about *micro* that we wouldn’t have seen otherwise, or point out a useful place to look, just as easily as the reverse.


b. 09.18.09 at 2:22 pm

“The appealing idea that macroeconomics should develop naturally from standard microeconomic foundations must be recognised as a distraction. ”

Throughout the entire fresh’n’salty micro-marco of recent weeks, it took me until this sentence to realize just what these guys were selling. I have never had Econ 101 or any exposure whatsoever, but I did physics for quite a few years.

Let me rephrase: “The appealing idea that cognitive science should develop naturally from standard quantum physics foundations must be recognized as an incredibly stupid way to obtain irrelevant results in the most arduous way possible.” (*)

Where are these guys going, and why are they in that handbasket?

(*) No, Hopfield, MFA etc. does *not* count. Think again.


wjd123 09.19.09 at 3:57 am


My formula is similar to one of those “you are what you eat” formulas. It’s no more than a rule of thumb. If it’s a false dichotomy or not remains to be proved. Let me flesh my formula out some more so you can tell me why. I worry about the ego thing. But let me start from the beginning.

I’m thinking of Durkheim’s idea that the catagories of the human mind are social constructs and arbitrary ones at that.

My problem with classical economists is that they posit human nature as a given and proceed from there. Anthropology would indicate that such assumptions are arbitrary.

Where is this given from which economists assume economic man will act?

Think of organic man as the id and society as the superego. From where comes the ego? Without the id and the superego it doesn’t develope in ways most of us can relate to. The superego is part of our socialization. It’s the moral and intellectual scoll that says this way and not that.

Break the law and you face the guys in the blue suits. Talk crazy and you face the the guys in the white suits. But they are not the final arbiters for an autonomous individual. One can practice civil disobedience, one can hold eccentric views, one can say enough to the superego.

So how self awareness? Does our self awareness naturally grow into maturity from some metaphysical seed or is it the end of a natural process. As fond as I am of metaphysical explanations I feel on firmer grounds with the empirical. There are problems with Durkheim’s social catagories concept but it doesn’t put and end to enquiry in the here and now as the metaphysical seed concept does. The ego would seem to be dependent on the id and the superego.

For instance, the feral child deprived of the stimilus of language for a number of years is incapable of learning any language. The language learning part of the brain shuts down unless there is stimilus, social stimilus, saying this and not that. However, not every tribe speaks the same language.

Isn’t this, not being subjected to the same stimilus and not speaking the same language, a stumbling block for the economic man of Adam Smith. His man was local and temporal. He couldn’t have functioned in the Roman Empire. He would have had too many scruples. His sociology of morals even his sociology of knowledge would have been different.

Today one could argue that both these sociologies are becoming more and more universal. That’s an empirical statement about social constructs and not a metaphysical statement about the nature of man, otherwise, universality would be a given.

Macroeconomists are free to model the most efficient way for an economy to return to equalibrium. They are not free to assume the same economic man for every model.

It’s not enough to say that we are more alike than different when it comes to human nature without showing why.

Personally every time I’m exposed to the editorial board of the Wall Street Journal I’m left wondering from what planet these people came. And we know that instinctally dogs won’t soil the areas around where they eat. Greenspan assumed the same about economic man and was badly mistaken.

There is no thread in a capitalistic society that ensures that social interests will be served by individual interests. Nor is there one in a communistic or socialistic society. The best we can do to ensure a “just right moment” is to get the incentives–which includes sanctions–right for a human nature, here and now, as it does exist in a particular society so social interests won’t go begging.

It’s my contention that in dealing with the problem of “Who is economic man?” there is no one ideal or empirical “human nature” that can be fitted for all people at all times.


Chris Dornan 09.19.09 at 6:29 pm

Yglesias has suggested that you avoid what look like contentious metaphysical claims and stick to pragmatic, methodological justifications for not trying to go all the way down.

I can’t get very excited about this distinction (I am not sure that it is a menaingful distinction), but I guess it may be important to some.


dsquared 09.21.09 at 7:48 am

To say “it works in micro but not in macro” is (to be very unkind about it) a little like a Creationist’s argument that sure, there’s micro-evolution but not macro-evolution.

more like saying that Boyle’s Law is important for understanding the behaviour of gases, but won’t really give you a workable weather forecast.


JoB 09.21.09 at 8:50 am

The difference between gases in Boyle’s law and those in workable weather forecasts are of the rather obvious kind, the difference between an idealized an environment and a real one.

Why are economists so startled that their idealized contexts fall shott of actual predictions?

Why do they use the prefix “micro-“, as if the ideal theories are good predictors of what actually takes place on a small scale?


John Quiggin 09.21.09 at 9:58 am

I don’t intend any contentious metaphysical claims, just the point that, as in lots of contexts, models that work well at one level may not by the appropriate basis for modelling at another level (a point that goes both ways wrt micro and macro).


dsquared 09.21.09 at 11:29 am

I think Steve L makes a sensible point; in actual fact, microeconomics (in the sense of “price theory”, which is the sense people mean when they are talking about microfoundations) doesn’t have particularly strong microfoundations. Even if one doesn’t want to jump feet first into behavioural economics (and I don’t), it has to be said that all of the strategic behaviour of firms, planning and financing decisions, formation of tastes and new markets, path-dependence, etc ad nauseam, which is clearly observable in actual microeconomics (as in, the study of specific individual decisions), doesn’t obviously give you confidence that the aggregate is going to look much like a constantly clearing market outcome with a representative agent.

I seem to remember that some people had a go at coining “mesoeconomics” to describe Chicago-style price theory once, but it didn’t take off.


Tim Worstall 09.22.09 at 10:09 am

“more like saying that Boyle’s Law is important for understanding the behaviour of gases, but won’t really give you a workable weather forecast.”

Or perhaps that if you try and create a workable weather forecast without both understanding and incorporating Boyle’s Law (the relationship between pressure and volume is fairly important for the different layers of the atmosphere) you’re not going to get all that workable a weather forecast.


Chris 09.22.09 at 8:42 pm

@49: I don’t think that is actually true. You can observe that conditions X tend to be followed by weather events Y with probability Z even without any understanding of why it works that way — in fact, ideally, your attempt to develop a theory of why it works that way would come *after* you have confirmed empirically that it does in fact work that way.

Since humans, scientists especially, have a drive to understand things, you probably wouldn’t just sit there with a giant table of preconditions and outcome probabilities with no attempt to systematize them, but you *could* get workable weather forecasts out of such a system, to the extent that weather obeys any patterns at all.

You would probably end up discovering Boyle’s Law or something like it even if you hadn’t known it in advance.

Comments on this entry are closed.