macroblog

« How has the financial system changed? (And what to do about it) | Main | Is offshoring behind U.S. employment's current problems? »

April 18, 2011

Can Keynesians be anti-Keynesian?

Follow any policy debate, and you are sure to find a list of economists who support or inspire those on both sides of the issue. In The Economist, we find some of those on the roster for the new Republican leadership in the House of Representatives, and why:

"When Republicans proposed slashing billions of dollars from federal spending this year, Democrats circulated predictions by economists that jobs and growth would be hit. John Boehner, the Republican speaker in the House of Representatives, countered with an economic expert of his own: John Taylor of Stanford University. 'Nothing could be more contrary to basic economics, experience and facts,' Mr. Taylor asserted on his blog, which Mr. Boehner cited. By cutting government spending, he said, the Republicans would 'crowd in' private investment and create jobs.

"… if there is one ideology that unites today's Republicans, it is Keynesianism, whose nefarious influence they are determined to stamp out. 'Young Guns,' the book-sized manifesto of Eric Cantor, Kevin McCarthy and Paul Ryan, leading Republican House members, devotes several pages to the evils of Keynesian activism and its exponents in the administration."

One of the interesting things about the article is that among the economists cited as being among the critics of "Keynesianism," you find the names John Taylor, Robert Mundell, and Kenneth Rogoff. I find that list interesting because if you follow the links I attached to those names you will find work with models that are decidedly Keynesian in structure. Works by Taylor and Rogoff are, in fact, seminal contributions to the "New Keynesian" paradigm that dominates macroeconomics today.

As far as I know, none of these men have repudiated the basic worldview that motivates the referenced work. In fact, as recently as last year John Taylor approvingly described, as he has many times, a key characteristic of the paradigm for monetary policy that was in place the decades before the financial crisis:

"… the central bank has a strategy, or rule, to adjust the interest rate depending on economic conditions: In general, the interest rate rises by a certain amount when inflation increases above its target and the interest rate falls when by a certain amount when the economy goes into a recession."

I added the emphasis to the last part of that passage as it is a feature of the so-called Taylor rule that is entirely built on the foundation of the New Keynesian model.

How, then, to explain the Keynesian predilections of the economists mentioned as presumed carriers of the anti-Keynesian mantle? The source of the confusion, I think, goes back to the historical, but somewhat obsolete, distinction between so-called Keynesianism and monetarism. The latter was, of course, personified in Milton Friedman and his dispute with what was the orthodoxy in the three decades following the Great Depression. Lost in the early-days labeling, however, was the fact that the disputes were more about the empirical details of theory rather than the theory itself.

In particular, Friedman did not deny the effectiveness of policy in principle but rather its wisdom or impact in practice. This sentiment is exactly the one he expressed in his prescient and transformative 1968 presidential address to the American Economics Association:

"In the United States the revival of belief in the potency of monetary policy was strengthened also by the increasing disillusionment with fiscal policy, not so much by its potential to increase aggregate demand as with the practical and political feasibility of so using it."

In a recent essay on Friedman's views about the ineffectiveness of fiscal policy, Tim Congdon notes Friedman's views on the issue:

"Friedman offered two informal theoretical arguments for the virtual irrelevance of fiscal policy, as he saw it. The second was that fiscal policy is much harder to adjust in a sensitive short-term way than monetary policy. But the first was the more telling and deserves detailed discussion.… In Friedman's words, 'I believe it to be true… that the Keynesian view that a government deficit is stimulating is simply wrong.' The explanation was the wider effects of the way the budget deficit is financed. To quote again, 'A deficit is not stimulating because it has to be financed, and the negative effects of financing it counterbalance the positive effects, if there are any, on spending.' "

Though Congdon emphasizes different channels (associated with the mix of monetary and fiscal policy associated with deficit spending), those who follow such things may recognize in Friedman's remarks the notion of Ricardian equivalence:

"This is the idea that increased government borrowing may have no impact on consumer spending because consumers predict tax cuts or higher spending will lead to future tax increases to pay back the debt.

"If this theory is true, it would mean a tax cut financed by higher borrowing would have no impact on increasing aggregate demand because consumers would save the tax cut to pay the future tax increases."

My point is not to dispute or defend the truth of the Ricardian proposition. My point is that it has absolutely nothing to do with whether one believes (or does not believe) that the New Keynesian framework is the right way to view the world. The essential policy implications of the New Keynesian idea (like the old Keynesian idea) is that changes in gross domestic product can be driven by changes in desired spending by households, businesses, foreigners, and the government in sum. You can believe that and still believe in fiscal policy ineffectiveness, as long as you believe that total spending is unaltered by a particular policy intervention.

There are, of course, plenty of arguments against fiscal policy activism that do not require adherence to Ricardian equivalence, in total or in part. The most obvious would be the position that any short-term rush from stimulative policies is more than reversed in the long run by the negative consequences of higher tax rates on productive activity, or the redirection of private investment to lower return public spending. Again, the point is that a self-professed adherent to a Keynesian reality need suffer no doubts about the coherence of his or her intellectual framework if he or she objects to fiscal policies aimed at juicing the economy through greater government spending.

This whole discussion may seem like a bit of inside baseball, and perhaps it is. But the stakes in this debate are high, as clearly illustrated by today's announcement from rating agency Standard & Poor's that it reduced its outlook to negative on the triple-A credit rating of the United States. In my view, productive discussions about the truly pressing issues of our day are unlikely unless we understand where the disagreements lie—and where they do not.

Photo of Dave Altig By Dave Altig
senior vice president and research director at the Atlanta Fed



April 18, 2011 in Deficits, Federal Debt and Deficits, Fiscal Policy | Permalink

TrackBack

TrackBack URL for this entry:
http://www.typepad.com/services/trackback/6a00d8341c834f53ef014e6108ba6f970c

Listed below are links to blogs that reference Can Keynesians be anti-Keynesian?:

Comments

I'm a rank amateur, so please let me know where I go wrong.

Fiscal stimulus is government spending, and government spending comes from either taxes or borrowing (bonds). Since raising taxes is pretty much the worst possible policy in a recession, let's assume the stimulus is entirely borrowed.

Bonds are repaid via future taxes, hence the idea of Ricardian equivalence. However there's always uncertainty in the future. To the extent the public believes the recession is due to a temporary market failure which the stimulus can repair ("nothing to fear but fear itself")... to the extent they believe the stimulus may be offset by future spending cuts... Ricardian equivalence will fail. So 0 < R < 1.

Now let's compare to monetary stimulus. It's also funded by the sale of bonds* but this time the public sells them to the Fed. Except all bonds originate at the Treasury, and additional demand from the Fed must ultimately cause additional supply. This is especially likely since recessions often cause a flight to safety, sending organic bond demand high. And, primary dealers are required to sell.

So the only difference is whether the Treasury sells bonds to the public or to the Fed. The Fed destroys bonds and even refunds the coupons, so R = 0. The cost of the procedure is inflationary pressure. And inflation... also stimulates the economy,** by spurring investors to renew existing contracts and seek higher nominal returns. So Friedman was right, though maybe not as right as he thought or for the reasons he gave.

That's the financing issue. What about Taylor's crowding out? Here I think there are more teeth: http://lumma.org/microwave/#2010.07.18

-Carl

* Unless the Fed has been doing something weird, like you know, buying MBS. Those are money-quantity neutral if the payments are destroyed. So I don't get all the talk about an exit strategy. The Fed can simply hold its MBS to maturity, alternatively destroying or reinvesting the payments as part of its overall operations.

** Provided inflation expectations remain anchored. I think.

Posted by: Carl Lumma | April 18, 2011 at 11:03 PM

Essentially there is no difference between Keynesianism and its opponents like monetarism. They all agree on the basic notion that money is equivalent to cash balances (though how you can define money as something that is not spent and then try to relate it to total income or total expenditure is beyond me).

Keynes argued about the tension between bonds and cash balances. Friedman added stuff like equity and consumer durables and Keynesians like Tobin did not disagree.

Keynes's classic money equation
M = M1 + M2 = L1(Y) + L2(r)
where M2 is the speculative demand for money
simply reduces to monetarism if speculative demand is zero.

Posted by: Philip George | April 19, 2011 at 04:59 AM

Both Carl and Philip make good points. Carl's contribution is noting "uncertainty". Ricardian equivalence doesn't hold for, at least, two reasons: 1) the future is irreducibly uncertain, and 2) most people outside of finance and economics recognize this basic fact and thus heavily weight the present observable economic activity from employment-producing fiscal stimulus and assign very little weight to the distribution of possible future tax rates. Thus they don’t save their current earnings to pay the tax man. If you believe that people use future tax rates in their present consumption calculus, please provide me the probability distribution and parameters that they are using for their calculations.

Technically, Keynes posited that inflation stimulates the economy by increasing people's expectations of the marginal efficiency of capital (i.e., the return on invested capital). When capital is plentiful the marginal efficiency is low and people are not motivated to invest it in new capital projects that employ people and resources. This leads to the similarity with Friedman. People compare the marginal efficiency of capital to the schedule of interest rates when they decide to invest (notice that marginal efficiency of capital is not always equal to the interest rate as Classical, Neo-Classical, Monetarists, Neo-Classical Keynesian Synthesis [in the long run] schools all assume). Thus, when the marginal efficiency of capital (return on invested capital) is greater than the interest rate people invest, when it's not they don't.

So, there are 2 options, in general: raise the marginal efficiency of capital through boosting confidence ("animal spirits") and/or use an “exogenous” force to employ people to build capital (e.g., fiscal stimulus). Keynes also believed that you could lower the interest rate below the schedule of the marginal efficiency of capital by increasing the money supply as Friedman would advocate. However, Keynes believed this was inefficient because there is no substitute for money so as people demand more and more in uncertain times as they struggle to build a security blanket made of dollars big enough and they just let them sit in bank accounts, under mattresses, or in Treasuries (Keynes’s used the word “hoard”). This idle money cannot increase employment (Keynes's main goal for moral and economic reasons).

This is exactly what we have seen--corporations haven't really invested and banks haven't lent (due to lack of willingness and now they claim lack of demand) thus it's left to the government to spend and increase employment. One might point to increases in the equity markets as a mechanism for saved money to go towards employment producing investment. But, remember that an overwhelming amount of money that has flowed to asset markets post the crisis was in the secondary markets and not primary issuance (e.g., IPOs). Thus, the money is simply trading hands based on past investments that employed people to build the products or provide the services of the businesses represented by the ownership stakes provided by stocks. Outside of employing people in the financial community, this “spending” doesn’t employ many people. This is why you see a tight labor market for people in finance and a weak market for construction workers.

Posted by: Joshua Packwood | April 19, 2011 at 07:14 PM

Keynesians are anti-keynesian. Monetarists are keynesian. What a confusion!
Have any of you ever tried marxist economic theory for a change?

Posted by: wskarma | April 20, 2011 at 04:56 AM


Thanks, Dave for a thought-provoking and brave post. A few comments. First, neo-Keynesianism dominates the world of applied macroeconomics because it has empirical content. The theory can be tested and implications for policy made based on parameters derived from experience. True tests come when events occur outside the historical norms, such as what has occurred in the past few years, and in general the theory has stood up well.

As a result, this paradigm dominates the applied world, where there are quantitative benchmarks for performance. Unfortunately in other realms the theory does not dominate, and in fact is ridiculed. The first such realm is academia. Here, neoclassical theory has all but squeezed the neo-Keynesian paradigm from the classroom. The grip of the neoclassicals is such that one can barely find a neo-Keynesian article in a major journal. The benchmark in this world now is model elegance. Abstract optimization is prized far over empirical validity. The fact that these models rely on unfounded assumptions such as rational expectations and efficient markets and that recessions are the result of productivity shocks are not subjects for discussion. No $100 bills on the sidewalk, but trillions in losses.

The second realm is policy. If we give conservatives the benefit of the doubt as to their motives (a significant assumption), then what they feel is a strong aversion to government action of nearly any type. This explains the mental gymnastics of the economists you cite whose research has been based on an empirically valid theory, but whose implications run into their ideology. The resulting cognitive dissonance is an embarrassment to them and to the profession. I put it this way: a Keynesian would say that the government should pull the Chilean miners out of the mine, while a conservative would say it's not effective because either it would encourage future miners to be relaxed on the job or because the cost would imply that the private sector would be crowded out and less money would be spent on safety.

One last point: your last line on S&P and the US debt rating. Of course, the problem is serious and needs attention. But please do not encourage these guys by citing them as a source of authority on these matters. Their track record on sovereign debt is abysmal, to say nothing of securitized products. Their incentives are highly questionable and their use by investors as a crutch is a key factor behind the financial collapse we have just gone through. Please devote your efforts to unwinding them and encouraging independent analysis.


Posted by: Rich888 | April 20, 2011 at 09:56 AM

I believe the concept that, with a Federal deficit, one would spend less now because taxes are going to be higher in the future is just not in the American conciousness.

Indeed, I see as the other way around: people pulling income into these low tax years, and then increasing spending now, rather than have the income in what might be higher tax years in the future.

But this is an empirical question. Surely polls have been done: "Are you spending less now, because taxes might be higher in the future?" To which the answer, "What are you talking about?" should be a possible choice.

[this comment occured because of a Krugman link]

Posted by: David Fields | April 20, 2011 at 09:59 AM

David Fields has it correct. The future for 70% of Americans is the bill due at the end of the month, not the marginal efficiency of capital.

Posted by: DR | April 20, 2011 at 11:34 AM

I would think that David Field's comments on testing Ricardian equivalences with the American public might be applied to corporations as well. If corporations were concerned about future taxes, we would see that factored into the tax rate used in DCF analysis for longer-term projects. There would be increased tax-rate assumptions in later years, or at least future tax rates would be factored into the sensitivity analysis.

It would be easy enough to poll corporate finance managers to find out if this was in fact happening, and then try to calculate what impact that had on the acceptance of business projects. My suspicion is that it might have an impact, but at much less than Ricardian equivalence.

John

Posted by: Ragweed | April 21, 2011 at 03:06 PM

"If we give conservatives the benefit of the doubt as to their motives (a significant assumption), then ..."

If you are going to claim to give conservatives the benefit of doubt as to their motives, then you should do so. You do no such thing. It would be as if a conservative said: "If we give liberals the benefit of doubt as to their motives, then we should assume [the same exact things we already assume without giving them the benefit of doubt]." If you are unable to accurately or objectively describe the positions and motives without your own biases, try not to claim that you are in fact doing so.

Posted by: sparky | April 22, 2011 at 07:36 AM

Keynes did not ponder what drove the economies of the world in order to get published or to achieve tenure. He eschewed elaborate mathematical economics, even though his intellect and mathematical skill fully permitted him to engage in such analysis. Rather, he was practical, and looked for an explanation in the Great Depression of why what he observed was as it was. A highly successful investor, he spent the first hour of every day doing the equivalent now of reading the WSJ and the Financial Times. He also had worked for the British treasury. He was directly responsible for financing the First World War for Britain. From these practical, real life experiences, he arrived at economic conclusions that are indeed the paradigm by which most operate today, including Mr. Taylor. So, the Young Guns have cited economists who have stood on the shoulders of Keynes.

Last, we make a grave policy error if we fail to recognize that the steps of governmental intervention taken by Chairman Bernanke, the Congress, and Secretary Paulson, in the early fall of 2008, are the chief reason that we did not see in late 2009 and early 2010 unemployment go to 20%. Forgotten, for example, in the current discussion of cutting government spending, is the effect the stimulus package had on state governments, which would otherwise have severely curtailed expenditure. The federal government used its collective strength to avoid the kind of reamplification of recessionary tendencies that would have been the inevitable result of such state layoffs.

Finally, it does not help to analyze these economic issues in terms of conservative and liberal, or Republican and Democrat. As the collapse of communism demonstrates, rigid adherence to a purely ideological notion of economics is doomed to fail. Rather, the policy issues should be addressed in a practical way, avoiding dogma and with a keen eye to the demonstrable.

Posted by: Michael Egan | May 24, 2011 at 10:14 PM

Post a comment

Comments are moderated and will not appear until the moderator has approved them.

If you have a TypeKey or TypePad account, please Sign in