Democracy in Deficit and holding A Tiger By the Tail

|Peter Boettke|

Since 2008, I have argued that to gain an understanding on the current situation economists should consult James Buchanan and Richard Wagner's Democracy in Deficit and Hayek's A Tiger By the Tail.  Its not like these are esoteric references, but instead summary statements from 2 of the 3 top critics of Keynesian economics among Nobel Prize winners (Hayek, Friedman, and Buchanan).  Buchanan tackles the political legacy of Keynes as embodied in deficit financing and accumulating public debt.  Hayek addresses the inflationary legacy of Keynesian economics.

Current affairs seem to confirm the Buchanan and Hayek analysis.  Manipulation of money and credit (evidenced by deviations from the Taylor Rule) resulted in a pronounced business cycle culminating in 2008.  Debt crises plague not just Europe, but at the municipal, state and federal level in the US.  And both the ECB and the Fed have engaged in significant efforts to meet the economic crisis with aggressive monetary policy.  So aggressive in the US, in fact, that monetary policy has been focused not only on short term manipulation of credit (QE3), but also efforts to make sure that long term signals that spur inflationary hedging are muted (Operation Twist).

However, fewer economists than I would have believed take the warnings of Buchanan and Hayek seriously especially given the political and economic history of the past 60 years.  This is true even among many market-oriented thinkers.  See, for example, David Beckworth's recent post at Macro and Other Market Musings "The Biggest Myth About the Fed," and Scott Sumner's recent post at The MoneyIllusion "It makes very little differnce how new money is injected."  

I'd like to see an honest and thorough debate on these issues.  At a fundamental level I think the debate isn't empirical but analytical, and a consequence of the very Keynesian style of theorizing that Hayek and Buchanan wrote against.  It is just very difficult for economists trained since 1950 to think consistently in a framework that does not permit "macro" aggregation.  Or, as Roger Garrison has put it over the years, 'While there may be macroeconomic problems (inflation, unemployment, business cycles), there are only microeconomic explanations and solutions." In other words, the only economics that works is relative price economics.  But the ability of economists to think consistently and persistently in relative price economics terms is difficult and it is easy to slip into the habit of aggregative thinking, such as price levels, etc.  The old habits of thoughts which stifile thought, as Keynes once wrote about classical economics, have now become Keynesian habits.  In all sincereity, I don't want to assert this as the definitive answer, but only as the beginning of the conversation.  But how else would you explain Scott Sumner's discussion of injection effects, or David Beckworth's discussion of monetization (btw, see last sentence to paragraph 2 for answer to his question about the empircs)?

As Hayek once said about Schumpeter in a different context, when habits of thought mislead even astute thinkers, it is time we seriously consider abandoning those habits of thought.  The persistent and consistent applicaiton of opportunity cost reasoning, and the focus on relative prices and the correponding adjustments that shifts in prices engender, provides a framework of economic analysis that must be adopted.  But just as the Keynesian transformation of economics required new data collection, and new data analysis tools, as well as new institutional environment within which the practice of Keynesian macroeconomics was conducted, so too will a wholesale shift in theory, empirics, and policy follow from a reawakening of economics along price theoretic lines.

Comments

[macroman1]

Beckworth also had this critical piece: http://macromarketmusings.blogspot.com/2012/11/the-fed-budget-deficit-and-facts.html Not sure what to make of it.

[Daniel Kuehn]

re: "The persistent and consistent applicaiton of opportunity cost reasoning, and the focus on relative prices and the correponding adjustments that shifts in prices engender, provides a framework of economic analysis that must be adopted."

I'm not sure we ever dropped this, Peter. It's always been central to my education, the instruction I've had the privilege of giving, my research, and my discussions with others. I think part of the problem with sorting out these disagreements is when the disagreement is misdiagnosed like this. You're not going to make any headway at all telling people they need to think about opportunity costs and relative prices. We will agree with you strenuously, and then we'll figure the discussion is over and then what you're interested in won't get covered. And that's a shame because there are genuine disagreements to be hashed out.

[Daniel Kuehn]

Point of clarification. You write:

"In other words, the only economics that works is relative price economics. But the ability of economists to think consistently and persistently in relative price economics terms is difficult and it is easy to slip into the habit of aggregative thinking, such as price levels, etc."

Do you mean "to think exclusively" here? Because I do think economists think consistently and persistently in terms of relative prices. It's hard for me to conceive of an economics that is not price theoretic. But if you want them to exclusively think in those terms, then of course there is a real departure of views.

[Peter Boettke]

Daniel,

I think the analogy to Keynes's "habits of thought" in classical thinking is spot on. Imagine an economics where Keynes never had traction and national income accounting never was developed seriously, and institutions such as the World Bank, IMF, let alone the Fed didn't exist. What would economics look like as a discipline today?

Confusion in science always results when people use the same words to mean different things, and different words to me the same thing. So I know I am risking confusion. But this is why I use the analogy to the transformation that economics took as a result of Keynes. Even market oriented thinkers I am contending are trapped in the habits of thought wrought by Keynesianism. This is why the insights of a Buchanan or Hayek are not fully appreciated and incorporated, though they are paid lip-service too.

See Buchanan's Cost and Choice for an examination of how economists while using the term opportunity cost have nevertheless not understood the elementary principles involved in the persistent and consistent application of opportunity cost reasoning.

I do realize I am asking a lot, but as I said I don't want to use this as the end statement, but instead as a spur for honest conversation on where economics goes from here. I hold an extreme position in this debate. But I am pointing to 2 recent posts by folks that are very intellectually sympathetic but who I believe go astray with respect to fiscal and monetary policy precisely because they are "slaves to defunct" habits of thought.

[Daniel Kuehn]

re: "Even market oriented thinkers I am contending are trapped in the habits of thought wrought by Keynesianism"

By "market oriented" I am assuming you are strictly meaning libertarians or Austrians, right? If so I think this is a good point. I have a very hard time taking critics of our use of aggregates (which I don't think are a bad habit of thought at all) seriously given the people they like to cite (from Hayek through Garrison - all of whom I think make important contributions) use aggregates as well.

Hayek is an interesting case. He was using aggregates before the General Theory or regular national income accounting came along. And he was following in the great tradition of Smith and the classics who also talked regularly in terms of aggregates.

I think economics ought to move in the direction of agent based work for a lot of the reasons Colander lays out. I imagine you're sympathetic with that. That's not the only way of doing relative price macroeconomics (a lot of standard macro models do that too). But macro is in the same rut that cosmology is in a lot of ways. Like cosmology, I think running simulations could be really helpful.

But I don't agree that that's because there's anything especially wrong with a lot of the macro gets done. I think we should do agent based modeling because I'm a methodological pluralist.

[Daniel Kuehn]

On second thought it might not be fair to say cosmology (or macro!) is in a "rut". Let's just say they face some of the same constraints in their work.

[Greg Ransom]

The debt crisis in at the state and city level in the United States is almost entirely a product of government worker unions & the explosion of compensation for unionized government workers.

"Debt crises plague not just Europe, but at the municipal, state and federal level in the US."

[Major_Freedom]

Daniel, there is a difference between paying lip service to relative prices and spending, and seriously integrating them into discussions of fiscal and monetary policy, the models and economists of which focus almost exclusively on aggregate prices and spending.

Aggregates dominate macro-economics, and macro-economics dominates fiscal and monetary policy models and discussions.

[Peter Boettke]

Robert Higgs is brilliant as always ... http://www.independent.org/pdf/tir/tir_14_03_10_higgs.pdf

[Alex Salter]

I saw in the comments section of TheMoneyIllusion that Scott Sumner explicitly rejects the idea of Cantillon effects. The theoretical gap might be even larger than it at first appeared. Obviously a different theoretical framework can render relative price changes as a result of money flows meaningful, but now we have to consider how we can bring about that kind of a paradigm shift. Empirical work on Cantillon effects seems extremely difficult, given the usual problems with identifying the counterfactual. How can we engender a paradigm shift when the evidence most likely to cause belief updating -empirical work, preferably econometric in method -isn't an option?

[geoffrey lea]

Hello, everyone.

Pete, I think one (admittedly weak) answer to your question is the combined first and second chapters of my dissertation. I didn't lay it out as an overt project of methodology (because that's the last thing I needed in my life: a pure methodology dissertation), but taken together, here's what I came up with:

1. Price theoretic macro is "microfoundations" macro, but it's not a reductionist microfoundations, like representative agent, but one that recognizes the institutional structures within which individuals make choices and those choices interact to produce macro results. This is Wagner 101 (aka Macro 701). It looks more like institutional economics than it does like conventional modern macro.

2. Our foundation would be microfoundations of individual choice in different institutional contexts, but in building a macro-structure that arises from this and sits on top of it, price theoretic macro resembles coordination Keynesianism, capital theory, and applied monetary theory. Put differently, it's the 3 L's: Leijonhufvud, Lachmann, and Lelandyeager (I cheated).

A little extrapolation: whether we like it or not, "macro" relies on indicators, and the definitions of the business cycles are determined in terms of macro variables. Falling real GDP is caused by drops in aggregate expenditures. The feedback mechanism is Say's Law (the Kates/Horwitz/Hutt version), which is--to my mind--eerily familiar to Leijonhufvud's story of "the corridor" and instability. What defines the corridor? The capital structure. Each individual and firm makes investments and contributions to the overall capital structure, but no one controls it. It's a spontaneous order (duh). The specific pieces that make up the capital structure are determined by individuals and firms, responding to relative prices, their own expectations, and the institutional climate in which they find themselves. As these variables change, the capital structure will change. Change "too much" or "too fast" and you might shake the whole thing; you've moved outside the corridor. Lastly, money matters. Monetary policy is itself an institutional feature; it's tightness and looseness matter. Money facilitates the process by which people make investments into the capital structure, but it also can distort the signals they they use to make those decisions.

[Allan Walstad]

Enthusiastic second on that Higgs essay.

[Greg Ransom]

Hayek had a single theme from his 1928 book to his writings and interviews at the end of his life -- math constructs *fail* to capture the imperfect *signal* function of prices and the imperfect learning, judging and discovery role of real human beings.

And a simple joke made by a senior LSE faculty member revealed the key pathology of those who mistake their mental constructs for the real world, the notion of "given data" -- ie the given, given.

Our mental constructs are made up of what we have stipulated as "given" -- these are perfect, set, fully known, fully surveyed, conceptually unchanging, entities and relations between entities.

But what is going on in the world is not a "given given", it lacks all of these properties, and our groping adjustments and our imperfect rough coordination of our doings with others also lacks all of these properties.

But, ultimately, our understanding of the terms and concepts stipulated in our "given given" mental constructs appeal for their public significance to these structures in the world which are *not like* the formal entities in the mental or math models.

Because we can talk of the entities in our mental models as if these models are populated by the structures in the world, we soon confuse the entities talked of in the mental or math construct for the real causal goings on in the world.

We reify the model, and misunderstand the world.

People need to understand that this is exactly the *same* insight that dawned upon Ludwig Wittgenstein -- mental constructs made of "given givens" which are given to one mind fail to capture the imperfect networks of imperfectly coordinated common patterns of orienting our self in the world, and these 'given given' entities mislead us about the source of significance or meaning -- we go looking for significance in 'given given' entites, eg Plato's essences, Russell's logical atoms, Frege's senses, Kripke's direct references, Lewis's possible world entities.

When we reify the mental construct we've built of 'language' using logic and meaning entities, we misunderstand the real world of the social phenomena of language.

In the economic case, we lose sight of networks of price relations in the world as found social tools already existing in the world for orienting our doings in coordination with other people. Social ools with aren't always perfect and don't always guarantee perfect coordination.

In the language case, we lose sight of networks of shared practices of going on in the world involving speech and written words as found social tools already existing in the world for orienting our doings in coordination with other people. Social tools with aren't always perfect and don't always guarantee perfect coordination.

Prices and language are external social network realities with an existence and reality far beyond the closed system of any single individual's formal mental model of the price system or the language system made of of 'given givens' or stipulated 'meanings' and their formal relations.

The lesson of Mises' socialist calculation argument and Wittgenstein's private language argument is that we can't recreate this social thing that lies outside us in a fully surveyed formal system of "givens", we can't recreate it and we can't replace it, what we do is make use of it coordination our social doings within this larger *socially* given network of relations, which we don't receive as 'given given' entities like a hat in a box, but as networks of significance we are constantly orienting ourselves within and internalizing, in the first instance without any real explicit articulated fixed rules of going on together in a coordinated way. We acquire usefully and commonly coordinated practices via imitation, trial and error, training, absorbing the culture, practice, getting advice from others, etc.

In short, the explanatory divide between Mises/Hayek and those limiting the conceptual space of their economic explanations to formal relations between "given given" in formal mental or math models is a deep one, and is shared with the chasm dividing the later Wittgenstein from the whole of the Western philosophical tradition as it approaches language, knowledge, and other core social phenomena.

It's a hard core conflict of vision.

But its not a conflict of visions without many great victories for the Mises/Hayek/Wittgestein side -- see Thomas Kuhn's social learning, background understanding achievement against the "given given" mental construct tradition in the philosophy of science.

[Bill Woolsey]

"Manipulation of money and credit (evidenced by deviations from the Taylor Rule) resulted in a pronounced business cycle culminating in 2008."

Really? Really?

The Taylor Rule?

A mathematical formula relating interest rates to the inflation rate at the output gap?

And you complain about aggregation?

No matter how wrong headed they are, all they have to say something that is even close to the Austrian Business Cycle and you overlook absurdity.

If anything, the Taylor Rule is responsible for the Great Recession. Targeting interset rates is always dangerous, and it has proven to be a disaster. Inflation targeting was never a good idea, and it has been proven a disaster.

[http://gy2012.in/home.php?mod=space&uid=66700&do=blog&id=252236]

True Religion outlet it was still a long way Jeans Outlet adventure they sailed on south and a little east for twelve days with a gentle wind,the skies being mostly clear and the air warm,and saw no bird or fish,except True Religion Jeans Outlet USA once there were whales spouting a long way to starboard.Lucy and Reepicheep played a good deal of chess at cheap True Religion Brand Jeans Outlet time.Then on the thirteenth day,Edmund,from the fighting top,sighted what looked like a great dark mountain rising out of the sea on their port bow.
http://gy2012.in/home.php?mod=space&uid=66700&do=blog&id=252236 http://gy2012.in/home.php?mod=space&uid=66700&do=blog&id=252236

[FC]

Now there's one for Wittgenstein to ponder.

[Peter Boettke]

Bill,

I erred and gave a misleading impression. I was trying to provide some evidence for those who are in fact trapped in the older habits of thought that empirically interest rates were in fact pushed lower than what inflation targeting would have advocated -- and thus produced a boom-bust cycle.

You are right to call me on the inconsistency. But I don't understand your claim that the Taylor rule is responsible for the Great Recession.

What I am really concerned about in this post is whether or not certain habits of thought systemically underestimate the costs due to inflation and deficit financing. I suspect so, and thus I would like to counter those habits of thought.

[Richard Ebeling]

If I may do my "usual thing," and refer to the comments of earlier economists, in this case on the relevance of the "injection" process in an inflationary process.

First Joseph Schumpeter from his "History of Economic Analysis (1954), where he pointed out that:

“The Austrian way of emphasizing the behavior or decisions of individuals and of defining exchange value of money with respect to individual commodities than in respect to a price level of one kind or another has its merits, particularly in the analysis of an inflationary process; it tends to replace a simple but inadequate picture by one which is less clear-cut but more realistic and richer in results.”

Schumpeter had considered the injection, non-neutral process to be important from an early time in his professional career. Thus, in his "Money and the Social Product" (1917):

“To begin with, increases in the quantity of money never occur uniformly for all people . . .

“Prices never rise uniformly – neither the prices of consumer goods relative to each other nor the prices of consumer goods relative to those of the means of production.

“Thereby the price rise ceases to be merely nominal. It means a real shift of wealth on the market for consumer goods and a real shift of power on the market for the means of production, and it affects the quantities of commodities and the whole productive process.

“No doubt, not all of these effects are permanent . . . But very frequently such re-establishment of the status quo is impossible.”

And for another internationally renown economist, also with "Austrian" roots, Oskar Morgenstern, from his 1972 article, "Thirteen Critical Points in Contemporary Economic Theory":

“The concentration on undifferentiated aggregates as, say, that of the total quantity of money, is a step backward into a more primitive world of thought. It runs counter to what must be done . . .

“Consider an inflationary, or as a matter of fact, any increase in the total quantity of money.

“If no account is given where this additional money originates from, where it is injected, with what different magnitudes and how it penetrates (through which paths and channels, and with what speed), into the body economic, very little information is given.

“The same total addition will have very different consequences if it is injected via consumers’ loans, or via producers’ borrowing, via the Defense Department, or via unemployment subsidies, etc.

“Depending on the existing condition of the economy, each point of injection will produce different consequences for the same aggregate amount of money, so that the monetary analysis will have to be combined with an equally detailed analysis of changing flows of commodities and services.”

(I might add that I took Morgenstern for the History of Economic Thought at New York University the last semester he taught before his death, and he several times during the lectures went out of his way to emphasize how the "Cantillon Effect" was unfortunately under-appreciated in modern macroeconomic theory, and therefore monetary understanding far poorer than it needed to be.)

Anyway, perhaps how the money is injected may make a significant difference.

Richard Ebeling

[Bill Woolsey]

The Taylor rule is responsible for the Great Recession for two reasons.

First, it enshrines interest rate targeting. When the target for the interest rate hits zero, what is to be done?

What a monetary regime really does is control the quantity of base money. If there is an excess demand for money, the quantity of money should rise to accomodate it regardless of the level of some particular interest rate, even zero.

Monetary policy is out of ammunition, we need a fiscal stimulus package? Whose fault is the explosion of debt? I blame the Taylor rule.

Second, is inflation targeting. During 2007-2008 there was a large run up in oil prices, leading to higher "headline inflation." The Taylor rule calls for responding to inflation with higher interest rates. Forcing down spending on output to slow inflation (or cause deflation) in other prices to offset the effects of a decrease in the supply of some particular product is always a bad idea. I think forcing down U.S. nominal income to keep consumer prices as a whole from rising due to a run up in the price of imported oil, even if caused by growing demand from China and India, is bad idea too.

Inflation targeting has other weaknesses as well. There is no error correction. Higher or lower inflation is allowed to shift the growth path of prices. This is bad when monetary disequilibrium has been allowed to fester, though a lesser of evils if there is a surprise shift in supply.

Oh, and the Taylor rule institutionalized 2% inflation.

OK, that isn't responsible for the Great Recession.

Under certain conditions, one can imagine that the Taylor rule would be a tolerable rule of thumb. But when conditions change, it can be a disaster.

The Taylor rule assumes that the natural interest rate is constant. It assumes that the relationship between "the" interest rate and the particular interest rate the central bank is targeting are constant.

Why should that be true?

Oh, and here is another problem with the Taylor rule. When the relationship between the interbank overnight loan rate and other interest rates--say BAA corporate bonds shifts, those wedded to this rule decide they need to fix all of the credit markets so that the past relationshp among interest rates holds. The risk premia that existed during the Great Moderation which is the period over which the Fed adjusted interest rates on average based upon past inflation and estimated output gaps and during which there were only two mild recessions and low inflation must have been right. And so, we need to have the Fed target credit interventions so that the differences between different interest rates return to those.

Why? It is the notion that the Taylor rule is right, and it is the markets that are wrong in how they price risk.

Taylor is a new Keynesian--one of the conservative Keynesians that have served as political advisors to moderate Republicans for decades now.

What bothers me most is how the old Monetarists have shifted over to his camp. I guess it is any rule in a storm. Just like the remarkably stable ratio of nominal GDP to M2 turned into an M2 quantity of money rule, the simple relationship between the Federal Funds rate and inflation and the output gap turns into a rule.

Oh, here is another problem with the Taylor rule. Sweep accounts allowed banks to avoid reserve requirements, and while I think reserve requirements are bad, they also made the M1 measure of the quantity of money worthless. The banks reported to the Fed checkable depostis after funds had been swept out.

MZM included lots of the things where the money was swept, which helped, though it is problematic because we know that many people hold things in MZM that are not useful as money.

Worse, money was being swept into things that were no longer measured anywhere. Overnight repurchase aggrements were dropped. And there was even overnight commerical paper!

Why didn't the Fed at least try to measure the quantity of money? They didn't care. They were targeting interest rates. And who said that was OK? Well, lots of folks, but the Taylor rule is a big justification.

When markets for overnight commerical paper and repurchase aggrements (using asset backed commmerical paper, with the assets being morgaged backed securities) collapsed, any complete measure of the quantity of money must of dropped. (The Divisa measures show this collapse.)

Oh, but we don't care about the quantity of money any more. We target interest rates.

Taylor rule.

As you can tell, I don't like the Taylor rule.

Oh, and PS. When I look at the data, there is a clear boom in the late ninties (the dot.com boom.) That is when interest rates were left too low for too long. But the Taylor rule says that instead interest rates were too low for too long a few years later. Well, only if you assume the natural interest rate--a market price--is unchanging.