Brian Snowdon, Howard Vane, and Peter Wynarczyk
A Modern Guide to Macroeconomics: An Introduction to Competing Schools of Thought
Aldershot, England: Edward Elgar, 1994. pp. 383-97.
AN INTERVIEW WITH ROGER W. GARRISON
How important do you think it is for macroeconomic models to have choice-theoretic roots?
Choice-theoretic roots are necessary but not sufficient. Explaining
economic phenomena in terms of the choices and actions of individuals is—or
should be—the primary business of economics. Nowadays this proposition
is almost as widely accepted among macroeconomists as among microeconomists.
It is simply no longer respectable to "let the data speak for themselves"
or to posit specific relationships among macroeconomic magnitudes while
remaining agnostic about the "transmission mechanism."
What are the key features of Austrian methodology which you would endorse?
The Austrians' methodological individualism—choice-theoretic roots,
as you say—is among the most important. Choices of individuals made in
the context of perceived opportunities and constraints are the basic building
blocks of the theory. Austrian subjectivism, which emphasizes the word
"perceived" in the previous sentence, is important too but can be pushed
too far. Shackle and Lachmann, who use the term "radical subjectivism,"
all but deny the existence of any underlying economic realities. It seems
to me that many aspects of Austrian theory involve a play-off of perceptions
What do you consider to be the key papers/books which have had a major impact on the development of your ideas?
The one book that stands out is F. A. Hayek's Prices and Production
(1935)—although I should say that by the time I first read it, I had already
read a lot of Mises and Rothbard and was ready for Hayek's triangles. Austrian
macroeconomics features the economy's capital structure and particularly
the time dimension of capital. Lengthening one leg of a Hayekian triangle
at the expense of the other represents a fundamental intertemporal trade-off
in which the creation of capital goods in temporally remote stages of production
requires a sacrifice of consumer goods in the current period. This capital-theoretic
analysis of the intertemporal allocation of resources impressed me early
on—especially in comparison to the conventional macroeconomic constructions
I was exposed to in my early graduate courses. There is just too much going
on within the economy's investment sector to be captured by a single aggregate.
How healthy is the current state of macroeconomics given the level of controversy? Do you see any signs of an emerging consensus in macroeconomics and, if so, what form will it take?
What is widely seen within the economics profession as the cutting edge
of macroeconomics is becoming increasingly divorced from economic reality
and policy relevance. Building models of the economy has come to be treated
as an end in itself. Insistence on mathematical methods and attention to
technique has virtually crowded out concern for whether or not the models
are actually "of the economy." The so-called Fully Articulated Artificial
Economies are all too often treated as vehicles for displaying some new
How do you view the rational-expectations revolution? Did it make a significant and meaningful contribution? Did it owe any debt to the Austrians?
The most significant positive effect of the rational-expectations revolution
has been to require macroeconomic theorists to make explicit their assumptions
about expectations. Before the revolution, all too many theoretical results
hinged on some critical but unstated assumption of systematic expectational
error. Sometimes simply articulating an assumption, for instance, that
workers take the cost of living to be constant when in fact it is steadily
increasing, reveals its implausibility. But if the word rational is stipulated
to mean consistent with—or, at least, not systematically inconsistent with—the
underlying structure of the economy, then the rationality of expectations
does not guarantee or even imply plausibility. How do agents know—or behave
as if they know—the structure of the economy? Adam Smith has taught us
that markets can work despite the fact that agents have little or no appreciation
of theoretical economics. All I'm suggesting here is that replacing know-nothing
agents with know-it-all agents is not always an improvement. We need to
theorize in terms of know-all-they-can-plausibly-know agents.
How would you classify the Austrian approach to expectations? If it is neither rational nor adaptive, then how is it best classified?
The Austrian treatment of expectations is guided by considerations about
what kind of knowledge market participants can plausibly have. Hayek often
makes use of the distinction between two kinds of knowledge. Theoretical
knowledge, or knowledge of the structure of the economy, is contrasted
with entrepreneurial knowledge, or the knowledge of the particular circumstances
of time and place. Economists have some of the first kind of knowledge
but not much of the second; market participants have some of the second
kind of knowledge but not much of the first. There is a certain formal
parallel, here, with the two kinds of knowledge (global and local) in typical
island parables as told by New Classicists. The difference between the
two constructions reflects a more general contrast between Austrian real-worldliness
and New Classical other-worldliness.
In your contribution to the Spadaro volume New Directions in Austrian Economics (1978) you provide a diagrammatical exposition of Austrian macroeconomics. What impact did this have within Austrian economics and did it lead to an improved dialogue with the Keynesians?
That piece had a very limited audience. Readers who were comfortable with the interlocking graphs didn't know enough about Austrian capital theory and Austrian monetary theory to make sense of them, and readers with a strong Austrian background were unwilling to cope with—and even were offended by—the graphs. I seem to remember getting some hate mail from people who thought that any attempt to express Austrian ideas graphically was a sacrilege to Mises. But there were a few people—mostly graduate students—who were receptive to both Austrian ideas and graphical analysis. For them, my exposition served its purpose. It demonstrated that all the individual pieces drawn from the Austrian literature fit together in an analytically consistent way. The market for loanable funds, the intertemporal structure of capital, and the inverse relationship between the rate of interest and the degree of roundaboutness all come together as key elements in the Austrian vision of the macroeconomy. These elements set the stage for demonstrating that monetary injections through credit markets lead to systematic but unsustainable distortions in the market process that governs the intertemporal allocation of resources.
Do you think that one can adequately model the Austrian approach given the emphasis upon uncertainty, historical time, and non-neutral money?
Well, my diagrammatical exposition was the best I could do. And admittedly, diagrams by their very nature tend to conceal the features that you mention. I can't imagine an Austrian model in the sense of a set of structural relationships that yield determinate values for current output, investment, production time, future output, etc. or in the sense of a Fully Articulated Artificial Economy. I think of a worthwhile model, though, as one that provides a stylized representation of the Austrian vision of the macroeconomy. It is not a substitute for verbal argument but rather a framework for formulating and presenting the argument.
Given your emphasis on the role of capital, what are the major significant contributions of Austrian economics in this area? Why do you think the work of both Hayek and Lachmann on capital and its structure have been generally neglected by both orthodox and not-so-orthodox (Neo-Ricardian/Sraffian) economists?
Two of the most fruitful contributions in the area of capital theory
were Hayek's Prices and Production (1935), which treated in the
simplest and most abstract form the notion that the economy's capital structure
has both a time and a value dimension, and Lachmann's Capital and Its
Structure (1956), which emphasized the extreme heterogeneity of capital
goods and the complex latticework of relationships among them.
What are the major strengths and weaknesses of the Austrian approach to macroeconomics?
Its weaknesses as well as its strengths derive from the capital theory
on which it is based. It is the focus on capital and interest that gives
the Austrian approach a certain directness and real-worldliness that is
missing in alternative formulations. For instance, when the central bank
initiates a monetary expansion, the new money enters the economy through
loan markets and therefore impinges initially on interest rates and in
capital markets. It has always struck me as odd that alternative treatments
of monetary expansion—I'm thinking here of the dynamics associated with
short-run/long-run Phillips curve analysis—so readily skip over this direct
effect and deal instead with labor markets and wage rates as affected by
asymmetries in perceptions and expectations. Key simplifying assumptions,
such as the one that replaces real-world monetary injections through credit
markets with the fanciful notion that new money is dispensed by a helicopter,
should have served as a red flag: something important has been left out
of account. Such fictitious constructions, I think, are designed to postpone—if
not avoid altogether—having to deal with the thorny issues of capital and
interest. But all too often the other-worldliness thus created is simply
accepted uncritically as the appropriate arena for macroeconomic theorizing.
Edmund Phelps has recently identified Seven Schools of Macroeconomic
Thought (1990) by distinguishing them in terms of the fixity or flexibility
of wages and prices and the nature of expectations about wages and prices.
I am inclined to lump all these theories together as macroeconomic schools
of thought that focus on labor markets and wage rates and contrast them
with the Austrian school, which focuses on capital markets and the interest
How close is Austrian economics to fundamentalist Keynesianism? Does the work of Shackle and Lachmann not demonstrate that there is an intellectual bridge between the two schools which may lead to cross fertilisation?
I think of fundamentalist Keynesianism as one of two extreme positions that help to locate the Austrian view as a middle-ground position. Shackle and, following him, Lachmann use the kaleidoscope as their model of asset markets: changes in the pattern of prices in asset markets are no more predictable than changes in the pattern of cut glass in a kaleidoscope. In stark contrast, hard-drawn versions of New Classicism treat the price mechanism as a clockwork rather than a kaleidoscope. The clockwork suggests an equilibrium-always approach; the kaleidoscope suggests an equilibrium-never approach. The Austrian view is somewhere between clockwork and kaleidoscope. Asset markets exhibit equilibrium tendencies but, because of the time element and the critical role of expectations, are particularly subject to disruptions. It is true that, like the Austrians, the fundamentalists emphasize such things as uncertainties, historical time, and the subjectivity of expectations. And these are the very ideas that are needed to jar the New Classicists away from their extreme position.
Streissler has suggested that Menger anticipated Keynes in emphasising the uniqueness of money among commodities. Would you agree? Would you also not agree that one can comprehend Chapter 17 of the General Theory more easily if one read, as a companion piece, Menger's classic 1892 Economic Journal paper on money?
Menger was concerned with the origins of money and with how money facilitates exchange. Forty-four years later, Keynes was concerned with the perversities of money and with how hoarding money can frustrate exchange. The "story of money" had a happy ending for one and a tragic ending for the other. It is true that both believed money to be unique and that Menger's saleability and Keynes's liquidity can be thought of as synonymous. I think that reading the two stories as companion pieces helps identify just where and how the Keynesian plot turns sour. For Keynes, the alternative to holding money is holding bonds—a view reflecting his belief that the decision about how much to save and the decision about what form the saving will take are made seriatim. The speculative demand for money, then, hinged specifically on speculation about movements in the interest rate. And the interest rate, according to Keynes, is not well anchored in economic reality. This construction led Keynes to psychological explanations of liquidity preference. For Menger, the alternative to holding money is any commodity for which money can be exchanged. Speculative demand—Menger didn't use the term—would have to reflect speculation on the part of the money holder that opportunities for making exchanges might present themselves. Menger, never mentioning the rate of interest even once in the entire article, had no reason to resort to psychological arguments and certainly never suggested anything in the way of a saleability fetish.
What do you think was Keynes's major legacy to economics?
From a practical standpoint, Keynes's legacy is the institutionalization
of demand-management policies. According to Marshall, prices adjust to
supply and demand conditions; according to Keynes, demand must be adjusted
to supply and price conditions. The Full Employment Act of 1946 provided
the mandate for government to make the appropriate adjustments. The monetary
and fiscal policy levers, as well as the system for collecting the data
that tell policymakers which way to pull on the levers, are tailor-made
for the Keynesian framework. As a result, the ideological leanings of the
policymakers don't count for much. The principle at work here is that those
in power tend to pull on the levers they find before them.
Had Keynes still been living in 1969 do you think he would have been awarded the first Nobel prize in economics? Would he have received your vote?
In the second and third years in which the Nobel prize was awarded in economics, the recipients were Simon Kuznuts (1970), who gave us the system of national income accounts and hence added the empirical dimension to the Keynesian research agenda, and Paul Samuelson (1971), whose "Keynesian cross" became the centerpiece of textbook Keynesianism. Many other recipients and even Milton Friedman, whose work on the consumption function was cited by the Nobel committee, have won the prize for their clarifications, extensions, or reformulations of Keynesian theory. It would have been inexplicable, then, not to have awarded the first Nobel prize in economics to Keynes himself. Would Keynes have had my vote? Well, Hayek and Myrdal shared the prize in 1974. Maybe Mises and Keynes should have shared it in 1969. I would relish reading the back-to-back Nobel lectures. (Incidentally, we might want to acknowledge that Tinbergen and Frisch, who actually did receive the prize in 1969, were both very worthy recipients.)
How important are Cantillon effects to the Austrian examination of the money transmission mechanism?
Austrian writers like to cite Cantillon as having been the first to
emphasize—in mid-eighteenth century—that inflation does not affect all
prices at once. Prices change in some sequence along with the spending
and respending of new money. During the process in which prices are adjusting,
relative prices are changing, and hence some quantity adjustments are occurring.
money is injected into the economy, then, may be as important than how
much is injected. This is what I mean when I say that the Friedmanian
helicopter should serve as a red flag signaling that something important
has been left out of account.
Are there any key lags in the Austrian business cycle approach which explain why such cycles occur and recur?
The focus on time and money and, more specifically, on capital gives
the Austrian theory a built-in lag structure. As I like to say it, "capital
gives money time to cause trouble." This aphorism summarizes a lot of Austrian
insights. Market participants have intertemporal consumption preferences,
which we take as given; the structure of capital, as guided by interest
rates and factor prices, will yield output in some particular intertemporal
pattern. It takes time to discover whether or not the intertemporal pattern
of output is consistent with the intertemporal consumption preferences.
A systematic mismatch, such as that created by an artificially low rate
of interest, will eventually become apparent, but by that time the economy
may be faced with the necessity of a major capital restructuring.
You have argued in the Journal of Macroeconomics (1984) that the two universals of macroeconomics are time and money. Apart from the Austrians, which other schools of thought handle these universals in an appropriate way?
We can find other schools that have promising approaches to one or the
other—but not both—of these aspects of the macroeconomy. For instance,
the time-to-build feature of some New Classical models has yet to be fully
exploited. Recognizing that an investment in Period 1 may influence the
decision to make some complementary investment in Period 2 brings capital
structure—or at least a hint of it—back into macroeconomics. Hayek entitled
one of his early articles, "Investment that Raises the Demand for Capital"
(1937) to emphasize the essential intertemporal complementarities that
characterize the capital structure. Unfortunately, the New Classicists
seem to limit the time-to-build concept to intra-firm complementarities,
which tends to trivialize it, and have introduced this concept in the context
of real rather than monetary distrubances. Note, though, that neither time-to-build
nor Hayek's title concept has a home in modern income-expendiutre analysis.
What are the strengths and significant contributions of non-Austrian macroeconomics since Keynes?
Apart from contributions I have already mentioned, I would say that
the application of Public Choice to policy analysis has been significant.
But it was Hayek, I think, who first offered a proto-Public-Choice theory
of inflation in his Constitution of Liberty (1960). In their Calculus
of Consent (1962) Buchanan and Tullock credited Hayek and went on to
apply choice-theoretic economics more broadly to political issues. The
built-in lags that separate the implementation of a policy and its ultimate
(negative) effects make for a natural blending of Austrian macroeconomics
and Public Choice.
Did the Austrian approach to business cycles adequately explain the interwar experience?
I think it is correct to say that the boom of the 1920s and subsequent bust is the clearest and cleanest illustration of the kind of intertemporal discoordination identified by the Austrians. The newness of the Federal Reserve and the general absence of Fed-watching even in financial circles allowed the monetary expansion to keep interest rates artificially low for an extended period of time. The Austrian theory, then, accounts adequately for the misallocation of resources within the capital structure—the malinvestment—during the upswing and hence for the inevitability of the bust. Other considerations must be added to the story to adequately account for the depth of the downturn and the duration of the depression. What Hayek called the secondary deflation is an income-constrained process in which the economy can spiral downwards far in excess of any needed liquidation. Milton Friedman's insights about the ineptness of the Federal Reserve throughout the 1929-33 period and again in 1937 come into play here. Also, the duration of the depression is to be accounted for largely in terms of the perverse effects of New Deal policies, which were well understood by the Austrians and others, but were not really a part of Austrian business cycle theory.
Given your work on Austrian business cycle theory, what insights do they offer to explain the behavior of the US and UK economies in recent years?
Let me confine my answer to the US experience. The boom and bust of
recent years is similar in form but different in particulars in comparison
to the boom and bust of the interwar period. I have made the argument that
while in the earlier period, policy-induced resource allocation was inconsistent
with underlying time preferences, in the latter period, policy-induced
resource allocation was inconsistent with underlying risk preferences.
In each instance, the focus is on the corresponding component of interest
rates—the basic rate of time discount for the interwar episode, the risk
premium for the recent episode. In the earlier episode, investment undertakings
were excessively long-term; in the latter episode, excessively speculative.
But in both episodes, the boom was artificial and hence the bust inevitable.
What are the key present research agendas of modern Austrian economics?
For several years, now my own research agenda has consisted largely
of putting capital theory back into macroeconomics. There is lots of room
for development on this front. I think that a well developed capital-based
macroeconomics can compare favorably with modern income-expenditure
models and with the New Classical parables.