Arnold Kling  

Blanchard on the Future of Macroeconomic Policy

PRINT
Central Planning in Health Car... The Problem with (Some) Experi...

Summarizing an IMF conference, Olivier Blanchard lists nine points. I will make extended comments below the fold. Here is a crucial sentence:


Monetary policy has to go beyond inflation stability, adding output and financial stability to the list of targets, and adding macro-prudential measures to the list of instruments.

Thus does over thirty years of academic thinking on monetary policy get thrown under the bus. Draw a cartoon with Scott Sumner lying under the wheel, screaming in agony.

Blanchard writes,


In the age-old discussion of the relative roles of markets and the state, the pendulum has swung--at least a bit--toward the state.

Are we talking about the broader public or about the elites? I think that the broader public might have felt that way early on during the crisis, but once the actual policies were put into place that began to change. I think that the public distrusts both markets and the state, and that is understandable.

Among elites, I think there was a lot of tendency toward closed-mindedness. That is, pro-market forces dug in harder and anti-market forces dug in harder. The way I see it, the anti-market forces were given a larger share of the bully pulpit at the IMF conference, which would account for the impression that the pendulum swung that way.

Blanchard continues,


The crisis made it clear that there are many distortions relevant for macroeconomics, many more than we thought earlier. We had ignored them, thinking they were the province of the micro-economist. As we integrate finance into macroeconomics, we're discovering distortions within finance are macro-relevant. Agency theory--about incentives and behavior of entities or "agents"--is needed to explain how financial institutions work or do not work and how decisions are taken. Regulation and agency theory applied to regulators is important. Behavioral economics and its cousin, behavioral finance, are central as well.

True enough. The phrase "agency theory applied to regulators" sounds like MIT-speak for Public Choice.

I am troubled by the ever-present implicit assumption that the knowledge problem does not exist. Blanchard writes as if technocrats can figure out where the markets are messing up and fix them. Actual experience be damned.

Paul Romer made the point that, if you adopt a set of financial regulations and keep them unchanged, the markets will find a way around, and ten years later, you'll have a financial crisis.
I wonder if anyone really wrapped their arms around that point. For my similar take, see The Chess Game of Financial Regulation.

Blanchard writes,


Pragmatism is of the essence. This was a general theme that came up, for example, in Andrew Sheng's discussion of the adaptive Chinese growth model. We have to try things carefully and see how they work.

That may sound innocuous. But Jonah Goldberg has warned me that pragmatism (the topic of his next book, if he ever finishes it) is dangerous as applied in politics. It means rulers using the ruled as guinea pigs for social experiments. When a firm conducts an experiment in the context of the market, it usually bears the consequences of bad outcomes. Not so much when you are spending other people's money. We have been conducting experiments using government money to fund "alternative energy" for decades, with no good results. See also the John Goodman piece cited on this blog by David Henderson.

My complaint about the conference is that it did not question some basic assumptions.

1. That there is something called the macroeconomy, which can be controlled using policy levers. As you know, I suggest instead that we interpret events in terms of the evolution of Patterns of Sustainable Specialization and Trade.

2. That technocrats can know enough to regulate financial markets successfully. Instead of seeing bad outcomes as a series of market imperfections, we might look at the interaction between market weaknesses and unintended consequences of policies. And, for whatever policy you propose, do not assume away unintended consequences.


Comments and Sharing





COMMENTS (12 to date)
B writes:

I am troubled by the ever-present implicit assumption that the knowledge problem does not exist.

Whenever this or Hayek are mentioned, accusations of being an Austrian will ensue. You'll then be chased out of the building by a mob. Thus ends all rational debate.

Philo writes:

Commentators on financial regulation sometimes deplore "regime uncertainty." They seldom notice Arnold's point about "The Chess Game of Financial Regulation": that there is no such thing as a satisfactory regulatory regime that could be put in place once and for all. "Financial regulation is like a chess game, in which moves and counter-moves proceed continually, eventually changing the board in ways that players have not anticipated," and that they could not possibly have anticipated. In practice, the players all know this; they know that "regime uncertainty" is an inevitable feature of financial regulation. To the extent that it is bad, regulation is bad.

Dan of the Fjord writes:

The knowledge problem is subsumed by incentives theory. So focusing on incentives will therefore deal with Hayek's "knowledge problem".

Cahal writes:

I was under the impression that thirty years of academic thinking on macroeconomic policy were already 'thrown under the bus' by the financial crisis?

Doc Merlin writes:

"Monetary policy has to go beyond inflation stability, adding output and financial stability to the list of targets, and adding macro-prudential measures to the list of instruments"

Sounds like the recipe for the worst monetary policy ever. Sounds like the fed will be completely captured.

fundamentalist writes:
Monetary policy has to go beyond inflation stability, adding output and financial stability to the list of targets, and adding macro-prudential measures to the list of instruments.

Hayek argued for this in the 1920's. The fixation on inflation did not begin 30 years ago. It began over a century ago. Macro has been stuck for over a century.

Hayek argued that monetary policy needs to move into the fourth generation by considering the effects of policy on relative prices, not aggregate prices. When the Fed pumps new money into the economy all prices do change in the very long run, but in the short to medium term it takes time for money to spread. As it spreads it distorts prices like a tsunami.

Since new money goes into spurring capital goods production first, it distorts the prices of capital goods, input and output, relative to the prices of consumer goods, which receive the new money last.

Like a tsunami, new money effects the people closest to the ocean first.

This distortion of prices sweeping through the economy causes booms and busts via the Ricardo Effect.

fundamentalist writes:
I am troubled by the ever-present implicit assumption that the knowledge problem does not exist.

That requires humility.

Les Cargill writes:

The issue is still the application of Bismarckian balance of power thinking to the "conflict" between State and Industry.

I can point to several otherwise intelligent people who will say "so what if regulation/the Fed/legislation creates rent seeking opportunities? We still need to do it for reasons of balance of power or the State will be eclipsed by Industry." And it's hard to refute - you can put up with some rent seeking if the cost of that is lower than not having what enables the rent seeking.

They don't beleive that Industry cannot capture government by market power - removing the negatives, they think Industry *CAN* capture government by market power. When you tell them there are (almost) no examples of Industry capturing government by market power, they point to various piracies.

The problem is that they ( those who say "So what?" ) do not see the costs. Or they see them and consider them acceptable because the so-whatters insert a "shadow column" in the accounting.

I conclude that people aren't interested - nor are they even familar with - *actual* fairness. They are interested in and familiar with activism as a method of expression of Hansonian expensive social signalling, they consider the balance to be worth it *to them* and cannot be dissuaded.

And so long as people who rent-seek call themselves capitalist, they'll be unshakeable and this will not change. The absence of rent-seeking may not be sufficient to get rid of preferences for activism, but it is probably necessary.

Ted writes:

Firstly, inflation stability (technically inflation expectations stability) is likely to guarantee output stability. Yes, if we are being technical about it, proper monetary policy requires very sophisticated behavior with regime switching along the equilibrium path between different rules and principles (it's been demonstrated now the Taylor Principle is neither necessary nor sufficient for optimal monetary policy and long-run equilibrium determinacy when agents have optimizing behavior along different equilibrium path and heterogeneous expectations). Although, for my own part, I've always wondered something. If the public isn't sophisticated enough to understand this sophisticated monetary policy behavior, wouldn't that screw with their expectations formation and lead to unnecessary stochastic output and inflation behavior? But anyway, for the most part, if the Federal Reserve is capable of stabilizing inflation / nominal income / nominal wage expectations - then this will obtain output stability. Or rather, I should say, the Federal Reserve wouldn't cause output instability. There is really nothing it can do about a purely RBC-type shock, like a news shock or if Kling prefers a recession from a productivity shock to certain sectors causing recalculation and mismatch. Blanchard is smart enough to know what I just wrote previously is true. I don't know why he is suggesting otherwise.

Secondly, in terms of financial stability, there is really not much monetary policy can do - at least monetary policy as we perceive it. What the Federal Reserve could have done in 2008 is that they might have seen the liquidity crunch, noticed the demand for safe assets, and done a portfolio rearrangement. They would sell off their Treasury holdings to satiate the demand for safe assets, then buy up some safe central bank's foreign bonds that aren't typically used by investors for liquidity holdings (like New Zealand) and hedged the portfolio risk using currency swaps. That would have likely helped quite a bit in panic mode of 2008 and could still maintain price stability (the Fed needed to commit to an explicit price level target in 2008 once the ZLB hit as well, this would have stabilized expectations, but that's for a different discussion). Some recent literature is suggesting including credit spreads in some interest rate rules might help, but this literature is still in its infancy - but it's effects will be too small to maintain true "financial stability." Beyond that, I'm not sure what monetary policy can do. The first policy is something to do once the crisis sets in and the second policy will only help us get ahead of the crisis a bit. Unless Blanchard meant something like central bank tools, which are a whole different bag of worms.

I don't like government regulations, but I think in the real world. I think the crisis of 2008 has sent a signal to big banks that now they can do whatever and they have a government backstop. This is dangerous. No commitment by our government to not bail them out will be credible. So, we have one of two scenarios. Either the commitment is followed, but not believed prior, and we get a huge crisis in 10-15 years as the banks thought they were protected. Or, more plausibly, they just get bailed out again in 10-15 years to a much bigger cost. That's why I favor something like taxing bank short-term debt, to correct the systemic debt externality. I also favor something like Zingales-Hart early-warning CDS system. Obviously though, I think the Dodd-Frank bill was a monstrosity that will suffer from regulatory capture, arbitrage, and government incompetence. Oh, and for anyone who thinks 2008 was "caused" by perceived guarantees, the evidence doesn't support that. Go look at major financial institutions CDS spreads in the run up to the crisis. The market definitely didn't believe the debt was guaranteed by the government.

fundamentalist writes:

Ted, Austrian business cycle theory has demonstrated that inflation stability does not lead to economic or financial stability.

Prices were very stable before the Great D and several other depressions. There was little before 2008. Hayek shows in his "Profits, Interest and Investment" that overall prices can remain very stable and still cause booms and busts because of changes in relative prices.

Often, high levels of productivity increases will mask huge price increases that should result form increases in money.

Once the depression has begun, monetary policy can't do much, but it can do a lot to prevent booms that lead to busts.

Scott Sumner writes:

Since I am also opposed to inflation targeting, I will refrain from "screaming."

Monetary policy cannot achieve financial market stability (as the Fed learned in 1929,) rather we need better regulation of the financial markets, such as abolishing the GSEs, requiring 20% minimum downpayments, limiting deposit insurance to $25,000, etc.

John Papola writes:

Hmm... so the guy who wrong in 2008 that the state of macroeconomics was good is... um... why are we listening to him again?

Robert Schiller said last year, when I asked "who regulates the regulators", that what Thomas Hobbs thought ideal was a monarch. I don't see how Blanchard is any different. He wants the unelected technocrats to have unlimited discretionary power and a mandate to make the economy perfect. That sounds like the road to serfdom to me.

Comments for this entry have been closed
Return to top