The Economy is Not a Sous-Vide Machine

Kevin Grier lets loose at Cherokee Gothic:

People! Check out this quote,

“Michael Gapen, chief U.S. economist at Barclays Plc in New York, said Fischer’s comments “reflect an ongoing divergence of opinion” at the central bank. Fischer “doesn’t see much room for running the economy hot” while Yellen’s views “seem to provide a wide-open door to do that. You have a chair and a vice chair who see policy differently right now,” he said.”

After the events of the great recession, it’s just amazing to me that people think the economy is a steak, the Fed is a precision sous-vide machine, and all we have to decide is medium-rare or well-done.

For the millionth or so time, the models implying the Fed can do this, completely and utterly failed during the great recession. There is also evidence that a large part of the good outcomes credited to the Fed during the great moderation were actually due to exogenous forces (i.e. good luck).

Neither the Fed nor the President “runs” the economy. There is no stable, exploitable Phillips Curve / sous vide machine that lets us cook at a certain temperature.

This Fed worship is more religious than scientific. The past 10 years should be enough to convince anyone with an open mind that the Fed’s power over the economy is quite limited and tenuous.

But I guess it’s comforting to think that the little old lady behind the curtain can fix things for us.

She can’t, Stan Fischer can’t, Bernanke couldn’t. Maybe the sous vide machine is unplugged?

Yup, whatever your prior was, after the events of the Great Recession, you should surely downgrade your belief that Fed has a lot of control over the economy and yet I see a resurgence of this view despite it being at all odds with the evidence.


I bet this belief resurges every election year.

More so 1) how long it's been since the last recession, 2) how well risk assets have done.

So Sumner gives us "The Midas Paradox" and you give us "The economy is not a sous vide machine guys!" Great rebuttal

That's what I was thinking. The Fed can't make things a lot better, but they sure as hell can make things a lot worse.

Boo, Tyler.

Oops... boo Alex.

Since centrally managing the money supply can only produce bad results (just as centrally managing the shoe supply or banana supply will only produce bad results), then yeah... The Fed can't make things much better, but it can make them a lot worse.

So, the decentralized management of the money supply by the shadow banking system: the big investment banks lending money market funds with AIG insuring the deposits with deregulated mortgage brokers churning mortgages on old home sales to feed cash to consumers while reaping high rents to the brokers funded by money market debt, basically a multi-trillion creation of money outside the control or even oversight of the Fed was a superior decentralized way to manage the money supply?

Who was the centralized money czar who told every money market fund, every bond buyer, to shrink the money supply drastically, by trillions, one week in October 2008?

" The Fed can’t make things much better, but it can make them a lot worse" [snip]


+1. Great link.

Economists are high priests, and the neo-classical Fed building in Washington DC is where they perform their entrail readings and sacrifices.

South Park, per usual, already on this:

If anything, we should now understand that economics only really works as a way of describing tendencies at an aggregate level. Trying to be deterministic with any specific event / policy / whatever fails.

Obama broke the Fed.

I remember that, he must have gotten a huge thank you from his madrassa.

When did Obama become president? Was it 2001, 2005, 2006, 2007, ..., 1981? Or does Obama have a time machine he invented and that he won't let anyone else use?

I'm reading now Arnold Kling's book "Memoirs of a would-be macroeconomist" and it's very good. It basically makes the claim that all of economics is a fraud. And this from a practitioner. It's also steeped in technical jargon that I like (I'm not an economist, but I like to read and study economics).

Sample blurb:

By late 1986, I am about to switch careers, getting away from macroeconomics altogether.  At this time, my biases are still in favor of Keynesian economics.  I am now out of step with mainstream economists in several respects:

1.  Mainstream economists now put rational expectations at the center of every model.  My instinct is that this puts too much emphasis on expectations about inflation and money growth as determinants of behavior.  I still prefer the “general disequilibrium” approach taught by Solow, which the academic community has not embraced.

2.  Mainstream economists have discarded the traditional Keynesian consumption function, in which consumers reflexively spend a given share of their income.  They have replaced it with a dynamic optimization problem, in which consumers shift their consumption in response to changes in interest rates.  I find this approach sterile.  Moreover, it seems to me to lead to predictions that are counterintuitive and not well supported by data.

3.  Mainstream economists reject macroeconometric models because of the “Lucas critique,” but they are developing methods for analyzing macroeconomic data that are robust with respect to that critique. I am not persuaded that these methods are helpful, given the other problems with macroeconomic data. I am more concerned with nonstationarity and sensitivity to specification.  In some sense, each quarterly observation of macroeconomic aggregates is unique.  There is not any statistical technique that can convert these observations into quasi-experimental data suitable for scientific analysis.

4.  Mainstream economists are inclined to simplify their models in many respects in order to keep them mathematically tractable under the key assumption of rational expectations.  For example, they reduce an economy to a single consumer/producer/entrepreneru.  I see this as imposing arbitrary straitjackets that keep theory disconnected from reality.  To me, the macroeconomics papers that appear in journals have all of the relevance of scholastic debates over how many angels can dance on the head of a pin.

It is worth repeating that in physics, researchers manipulate equations that are subject to empirical verification.  I have come to doubt the possibility of verifying equations in macroeconomics, and that in turn makes the manipulation of equations a less well-grounded exercise.   I forget who first coined the phrase “mathematical masturbation,” but I think it is very fitting.

Good stuff!

That is good. Sumner's book is good too, though.

I've been reading Sumner's repeated claims the Fed has been running extremely tight money since 2008.

He never exactly explains what the Fed should be buying, or how the Fed buying whatever existing asset he calls for them to buy would result in workers being paid to work.

He has also argued that selling real assets to buy labor goods is great for an economy like the US, and presumably Saudi Arabia. Again, no explanation how that economic theory eliminates the popular unrest of the unemployed, under employed, and dissatisfied employed.

I don't see how it will be OK when almost no American owns property in the US because we sold it all in exchange for imported labor goods from Asia. Saudi Arabia has been selling liquid land that others burn in exchange for labor goods that keeps Saudis unemployed, and that has not been OK.

The unemployment rate is relatively low now, so I don't think he's claiming the Fed could get a lot more of the currently unemployed jobs. He is claiming that the Fed has control over the money supply, and therefore over NGDP, and that we could avoid major recessions by maintaining the NGDP growth path. Australia does this while maintaining a 4% inflation rate, although he thinks we could manage with a lower rate.

As for what the Fed could buy, in his ideal world it's NGDP futures contracts. In the real world the Fed is currently constrained to buy federal debt, but in his view it's an arbitrary distinction. We're not really in danger of running out of federal debt to buy, but he sometimes brings up a hypothetical of the Bank of Japan buying every asset in the world as a counterexample to those who think it can't create inflation. If that were actually the case, the Japanese could all then retire, but obviously inflation would ensue long before then.

The problem is that economists prefer to deal in non-testable claims. Growth too slow - the Fed didn't do enough. Inflation too high - the Fed did too little. We need better tests.

The problem is economists have false impressions on how the natural sciences test theory, thinking everything in natural science is tested in a tiny experiment in a basement lab over an afternoon.

When I took my first college classical physics, the focus was on the precision of the data collection, and what we could conclude. We knew what the theory is, so we knew what the data should be, but if we turned in such data, we failed. We needed data that did not fit theory and then all the reasons for measurement error, then steps to take to measure the error and other steps to reduce such errors. Biology labs were far worse. Cutting up stuff and nothing looked like the pictures and so what we were doing in the lab seemed to refute the book.

The lab experiments were lessons in how hard it is to collect data to test theory. Later labs introduced the kinds of tricks and indirect measurements. I never took geology, but thanks to PBS and lots of reading, I see how data is collected covering billion year long "experiments". Astronomy is 99% data collection on billion year long experiments. No geologists, astronomers, and physicists have pooled their resources to create big bangs in the lab to test their theorit's.

Economists for the most part theorize that theory is proved without labor. But in the natural sciences, every theory is the product of massive investment of labor over centuries, and the constant problem is finding money to pay for lots more labor.

How about economists call for increasing the budgets of census, bls, commerce, treasury to collect five times as much data on the economy?

1. People need to believe in something
2. The Fed having fine-grained control over the economy is something
3. Therefore we believe!

I thought that most of the blogosphere, including Tyler Cowen, were coming around to the view that the Great Recession started in 2008 due to a monetary shock partially triggered by the Fed (they tightened just as the financial sector was melting down instead of easing heavily).

Of course the Fed didn't respond properly, but certainly US Fed policy has been better than that of Europe and it shows in our recovery. And the argument of the market monetarists is that a proper response would have turned 2008 into a mild blip.

Yes, the difference between the Eurozone (or countries pegging their currency to the Euro in hopes of joining) and the UK, Denmark, Switzerland etc seem to be sufficient to show that central banks & monetary police matter a whole lot with regard to the Great Recession.

What evidence do you have for the Fed Tightening?

Are you saying the Fed sent a memo to every money market funds manager and account holder to shrink the money supply by a few trillion?

I didn't get the Fed memo, but then I had been looking at the deregulated mortgage sector and expecting a crash worse than 1986-7 when banks were deregulated and lots of bad loans were being made.

I read the theory in the late 60s and 70s on how deregulating banks would reduce the number of zero bank crisis especially from 1935 to 1975 from excessive bank regulation to an even lower less than zero bank crisis. Since the virtuous bank deregulation to eliminate bank crisis, I've seen 3 big bank crises.

If a group of Ph.D. economists can't figure out the optimal price for cinder blocks or gasoline, how are they supposed to figure out the optimal price for public and private debt? Aren't they just facilitating moral hazard and fiscal irresponsibility at this point? We know how this ends.

The Fed should go back to being the banks' Bank, if not dissolved entirely.

Our belief in the value of paper money is more religious than scientific. Remind me, who has the printer?

you should surely downgrade your belief that Fed has a lot of control over the economy

You should surely downgrade your opinion in the Fed being able to effectively do its job. I'd argue that it's also "Fed worship ... more religious than scientific" to automatically assuming that the Fed was doing the best possible job with the tools it had, though. It's certainly consistent to say that the Great Recession has revealed that the Fed was not doing a good job, but argue that better tools and better action would have been superior.

Now, political constraints are real, even for the Fed, and good reasons not to put too much faith in supermen. A rule or model that cannot be followed by the Fed in a crisis because it is unpopular or faith gets lost when needed most is a problem. Yet in these comments I see much the same as the criticism of "lazy fairies" with regards to the Great Depression (even as economists would say that the Smoot-Hawley tariffs didn't help things and were proof that laissez faire, while a popular slogan, was hardly the real policy of the day, at least when things got tough.)

Yes, they don't have control but it is convenient that most people believe they have control. Then, a few placebo operations could influence market decisions and........results. That's the methodology, right? The only problem here is that the FED audience are not citizens but qualified investors. All people can read and understand how it works but never forget most of us are not the intended audience of the FED.

Page Scott Sumner.

Sumner's gonna be mad at Tabarrok for this one.

Of course, the problem with relying solely on monetary stimulus is that we can't inflate our way to prosperity. Sure, at a time of financial crisis, it's monetary stimulus that stops the collapse of asset prices and restores faith in the economy. But that's it. It's like that first martini: it feels great, but the second or third just defeats what the first one provided. Restore confidence, not excessive inequality. As I've commented many times, monetary stimulus and fiscal stimulus are alike in that both are redistributive, it's just that the former is redistributive upward while the latter is redistributive downward. There's a reason those at the top vociferously oppose fiscal stimulus. The Fed's critics are about as likely to make this observation as they are to suggest that inequality might, just might, be a source of our problem. Instead, we get a steady stream of nonsense, like raise interest rates now so we can lower interest rates later, or raise interest rates slowly now so we don't have to raise interest rates fast later. Tabarrok suggests that the Fed can't fine tune the economy. He's right about that, but so what. Fine tuning isn't the problem.

I could use at least a few references to assess this post. The economy might not be a steak, but straw men don't eat.

" models implying the Fed can do this, completely and utterly failed during the great recession".

No, just need moar.

"There is no stable, exploitable Phillips Curve / sous vide machine that lets us cook at a certain temperature."

Yes there is: moar.

The whole analogy is idiotic and everyone who takes it seriously should be banned from discussing this subject

"The curious task of economics is to demonstrate to men how little they really know about what they imagine they can design. To the naive mind that can conceive of order only as the product of deliberate arrangement, it may seem absurd that in complex conditions order, and adaptation to the unknown, can be achieved more effectively by decentralizing decisions and that a division of authority will actually extend the possibility of overall order. Yet that decentralization actually leads to more information being taken into account.
The Fatal Conceit : The Errors of Socialism (1988), p. 76 Friedrich Hayek, Nobel Prize Winner, 1974

Though I've never read anyone who claimed that Hayek was as clever a chap as Keynes, Hayek's thinking always seems to me to be much superior. Keynes's ideas contain too much flavour of Newtonian mechanics to be useful in Political Economy, as the discipline used to be known. Hayek's ideas seem to me to be shot through with the flavour of uncertainty principles, a much wiser approach.

Of course he is correct, but the FRB isn't designed to control the economy, is it? It is supposed to regulate the money supply. That is hard enough, but a few decades ago, Congress, in its never-ending lust to salve public opinion without taking any actual responsibility, mandated the Fed somehow prevent high unemployment, too. These are not goals that can necessarily be accomplished singly, but together they may even be in direct conflict - to the extent the Fed can act.

So we take a difficult task, make it impossible, & then complain it isn't performed well?

The Fed has control over the money supply. It does not have control over the economy. But it does have control over the politicians who make economic policy.

The Fed has either developed or stoked certain theories to justify its control over the money supply, and one of those is that it's the Fed's responsibility to manage the money supply in order to manage employment levels (despite never actually demonstrating the causal connection between the two).

These theories are just a front for what the Fed really does... which is manage the money supply. And it manages the money supply for the benefit of its member institutions. It has no other purpose.

The opinion of New York Fed chief William Dudley:

Actually, he says the economy is the steak. The Fed is the sous-vide.

The words you are missing are "real" and "nominal". The fed may not have much control over the real economy in general, but it has near absolute control on one degree of freedom of the nominal economy. The dollar isn't redeemable for anything else. It is worth what the fed chooses to make it worth. Equivalently, total nominal spending in the economy will be at whatever level the fed chooses to set. And if that choice is made badly it can damage the real economy.

Cue Tyler posting something on the wisdom of Scott Sumner in 1, 2, ...

We hit the zero bound in the great recession. We all learned in school that monetary policy gets wonky at the zero bound.

The stock and bond markets certainly seem to think that the Fed has a lot of influence over the real economy.

Actually I am confused by the claims of Austrian minded people like Alex and Kevin Grier. On the one hand they are constantly complaining about "malinvestment" caused by overly low interest rates, leading to "asset bubbles". OTOH they claim the Fed has no control over the economy. Well which is it? If no control, then there can be no bubbles surely?

The reality is, is that there is no socialist calculation paradox going on here when discussion monetary policy goals, as there must be a policy on money supply by the money supplier. Even if that policy is to maintain the stock of currency at the same level, or to reduce it, or to grow it at a certain percentage a year. You can't have a "no policy" by the Fed, it is impossible logic.

Given that we have to have a policy, then this follows there is an optimum policy for any given goal. If you seek to minimize unemployment in the short term, there is one policy that would work best for that, if you wanted to maximize growth over a period of 10 years, then there is another policy that is best for that. Of course maybe monetary policy is only a necessary, but not sufficient enabler for reaching your goal, but it doesn't change the fact that there is an optimum policy.

The reality is, is that there is no socialist calculation paradox going on here when discussion monetary policy goals, as there must be a policy on money supply by the money supplier.

Money--currency--is just the commodity in ultimate demand. Money, and future money in the form of credit, tracks the supply-demand curve like any other commodity. This is why a central committee has as much trouble determining the "optimum" supply of money as they would the optimum supply of paper cups. When the market wants more money, it has to dig it out of the ground and refine it, or grind coffee beans, grow tobacco, manufacture 5.56mm rounds, etc. The process isn't perfect, no human endeavor can be, but the need to calculate an ROI provides a natural check on monetary expansion.

By contrast, when the Fed wants more money, it just prints or digitizes it. This should be theoretically possible: just declare a legal tender and it is, to the satisfaction of everybody, fully backed by all the goods and services available for exchange. In practice, there are all sorts of perverse incentives: governments want revenue without politically unpopular taxes, investors want guaranteed returns, the new money is non-neutral in the short run, etc.

Again, if the Fed focused on being the banker's bank instead of the Wizard of Oz we could probably muddle along fine. But now they're trapped by their actions in 2008 and the distortions are showing up in government bonds. They are elsewhere, student loans and new apartments probably, but we won't really know until after the Bust.

So what do you want the Fed to do on the money supply - keep it constant at some arbitrary level, raise it? lower it? Link it to gold? Link it to paper cups? You can't have "no policy" on money supply, whatever you do is a policy.

Friedman said grow it at some "natural" rate of growth determined via econometrics. Doing that, and monitoring its member banks for solvency, would be a good, humble policy shift short of the politically unrealizable and perhaps, at this point, undesirable goal of letting the market supply its own money.

It's pure hubris to think that the Fed can guarantee "full employment" or determine where asset or consumer good prices should stabilize, much less the policy of printing money and buying our own debt with it.

AG - your humble policy is exactly what proponents of market monetarism are proposing. So you are in good company with people like Scott Sumner. So why do you think you are in opposition to people like me and him?

We agree more than we disagree; please review my comments. However, you seem to say there's no socialist calculation problem, and I say there actually is. It may be unavoidable for all practical purposes (lots of libertarian economists would disagree), but the "central committee" problem is there. My other issue with central banking is the perverse incentives which inevitably show up. Cheers.

Your argument is a presumption of the Fed.

Study market money.

Dr. Tabarrok,

Are you saying a central bank cannot provide as much stimulus as it likes, if sufficiently determined? How would you respond to the thought experiment that, if not, a central bank could buy all the assets in the world without hitting it's inflation target?

The Great Recession itself shouldn't cause you to downgrade how much "control" the Fed has over the economy, if "control" is interpreted as "impact." That is, if you believe monetary policy played a key role in the housing boom, then it's impact was very large indeed. The post Great Recession world has been quite anomalous, with negative rates, QE, private debt purchases and so on, and yes it seems that central banks cannot make the positive impact they thought.

Central banks will always have massive potential impact/control on the downside.

Got to be very wise with all these Fed discussion in order to be in safe scenario, I always keep it all straight forward which is easy and simple to do with OctaFX broker which is awesome with their luxurious features including small spread from 0.1 pips to high leverage up to 1.500 while there is also gigantic bonus which is up to 50% and is use able too, so that’s why I am able to trade so nicely and perform at best of my ability.

look at australia, the gov helicopter dropped 900 to everyone with a bank account and tax file number.. (it doesnt matter to the recipients whether it was debt funded or monetary funded... monetary would have been fine) no recession essentially.

negative ior and having money sitting in bank accounts doing nothing is obviously ineffective, how easy is it to see that?

just do a monetary helicopter drop to everyone with a social and lets see some spending, what are people going to save the money, when its worth more at t-zero than at t plus one, when some price rises kick in.

after we do that then you can tell me whether the fed is ineffective, all weve seen so far is that an ineffective fed is ineffective. trivially uninsightful analogy.

Comments for this post are closed