Scott Sumner  

The world's three smallest macro models

PRINT
The Long and the Short Runs... The Sum of All Fears...

In a comment over at TheMoneyIllusion, Britonomist pleaded for a transmission mechanism:

I'm sympathetic to market-monetarists, I'd love to be able to believe that problems like low aggregate demand can easily be solved using monetary policy alone even at the ZLB, rather than having to rely on politicians to enact policy to drive growth.

But when I see so many other economists and experts in finance point out again and again the unreality of assumptions used in many monetarist models, the extreme importance of banking and bank money & how banking is absolutely central to determining the broad money supply - I have absolutely nothing to counter them. I can mention vague ideas about alternative ways to boost broad money via portfolio-re-balancing, they can counter with all kinds of technical discussion to show this would never be enough, and because I've still yet to see a tractable model of the actual transmission from market monetarists, I'm completely unable to make a case for MM. If I at least had an explicit model of the transmission mechanism (taking actual economic reality into account, rather than just assuming the monetary policy = giving everyone more M directly) I could vouch for you guys properly, which is why I keep asking for one here.


In a recent post, Nick Rowe linked to a Paul Krugman paper that presented the "world's smallest macroeconomic model". At the end Krugman discussed what the model was missing:

What is wrong with this model? Don't get me started ... but actually there are three main objections that macroeconomists are likely to raise:

1. What happened to the interest rate? For most purposes we will want at the minimum a theory of employment, interest, and money; that means a model with bonds as well as money and goods, which means IS-LM. (See my note "There's something about macro").

2. More fundamentally, the quasi-static approach here is at best a crude approximation to a dynamic model in which behavior results from plans that are based on expectations about the future.

3. Finally, the output effects of money come from the assumption of price rigidity. Where does that come from? (Overwhelming empirical evidence, that's where - but why?).

All these objections help to set the agenda for the last six decades of research.

But if you are one of those people to whom macroeconomics always sounds like witchcraft, who is hung up on Say's Law, who cannot even comprehend how a shortfall of aggregate demand is possible - then the world's smallest macro model is a good place to start on the road to enlightenment.


Meanwhile, Nick provided an even smaller model. Nick also discussed the absence of a financial sector in his model:

It's got nothing to do with "too much saving". There is no saving in this model. It's a one-period model, dammit. Everything gets consumed during the period, then they die and time ends.

It's got nothing to do with the real interest rate being wrong. There is no interest rate in this model. It's a one-period model, dammit.


Yes, the essence of the business cycle is monetary shocks and sticky prices. Here's a third "model", also lacking interest rates: The Fed determines the value of base money by adjusting its quantity. Shocks to the value of base money change all nominal aggregates in the economy, since money is the medium of account. The explanation? Supply and demand. And once you change nominal total labor compensation, hours worked will change if nominal hourly wages are sticky. And when hours worked changes, output changes. That's it, monetary shocks and sticky wages. No need to bring in the financial sector.

If you want to add the financial sector, that's fine. But it won't help you understand 2008-09. The Fed caused NGDP and nominal total labor compensation to fall about 8% below trend, nominal wages were sticky, and hours worked fell about 8% below trend.

That's all. Everything else is a footnote.


Comments and Sharing


CATEGORIES: Macroeconomics , Money




COMMENTS (17 to date)
Don Geddis writes:

Nice post! Love the 3 tiny macro models. (BTW: First link is broken; missing a ":" after the "http".)

Britonomist writes:

I appreciate you taking the time to make a post to address a comment of mine directly, thanks.

I'm still not really satisfied unfortunately. I feel like a simple supply & demand of base money model doesn't adequately explain the crisis.

If the Fed caused the crisis through the base money channel, that means the Fed suddenly decided to drastically reduce the supply of base money - why?

Either that, or banks expected the Fed to set the amount of base money too low in the future, again why?

The final problem is that this model makes sense when there is a singular 'money' as the medium of account, but when bank deposits also become a medium of account themselves, and when the amount of bank deposits can fall independent of a fall in base money, this presents a problem for the model, no?

Thanks again.

ThomasH writes:

A footnote showing what the Fed did in 2008-09 to cause NGDP to fall below trend (and by implication what it should have done in 2008-09 to restore NGDP to trend) would be helpful.

I interpret you to be saying that the Fed did not do enough QE or do it soon enough or possibly not the right kind of QE, pre-announced as QE in an amount and of a duration that was limited only by the return of NGDP to the target trend. Is that correct?

Given that NGDP futures did not exist in 2008 what was the second best asset for the Fed to purchase?

Mr. Sumner,

How did the Fed cause NGDP and nominal total labor compensation to fall about 8% below trend?

Did the Fed reduce or threaten to reduce the value of cash by 8%, or increase it?

Possibly you could add some detail to the story.

Scott Sumner writes:

Don, Thanks, I fixed it.

Britonomist, You said:

"If the Fed caused the crisis through the base money channel, that means the Fed suddenly decided to drastically reduce the supply of base money - why?"

Not at all. It has no such implication. In late 2007 and early 2008, the Fed stopped increasing the base money supply. This caused NGDP to stop growing. This caused nominal interest rates to fall to low levels in late 2008. This caused the demand for base money to rise. The Fed didn't increase the supply of base money enough to keep up with the extra demand.

You said:

"The final problem is that this model makes sense when there is a singular 'money' as the medium of account, but when bank deposits also become a medium of account themselves, and when the amount of bank deposits can fall independent of a fall in base money, this presents a problem for the model, no?"

Not at all. These assets are not perfect substitutes, so changes in the supply and demand for deposits only matters to the extent that it effects the supply and/or demand for base money. Which the Fed can offset.

Thomas, You said:

"I interpret you to be saying that the Fed did not do enough QE"

That is not my claim. With the right policy (5% NGDPLT) no QE would have been needed.

The Fed should buy Treasury securities, that's all.

Andrew, See my previous replies. The value of money rose due to the Fed's tight money policy.

Britonomist writes:

"In late 2007 and early 2008, the Fed stopped increasing the base money supply."

But why?

"Not at all. These assets are not perfect substitutes, so changes in the supply and demand for deposits only matters to the extent that it effects the supply and/or demand for base money. Which the Fed can offset."

Wait, so what happens if there is a huge credit shock, banks suddenly become far more risk averse & refuse to lend to most new customers; meanwhile people start regarding many debt instruments as dangerous and their value starts to plummet affect peoples' and firms' net-worth/capital? How does the fed offset this with base money?

Philo writes:

"The Fed determines the value of base money by adjusting its quantity." And by creating market expectations about how it will adjust the quantity of base money in the future (?).

E. Harding writes:

"But why?"
-That's simple. The dollar price of gold set a new record and gas was staying above $3 per gallon.
"How does the fed offset this with base money?"
-QE until NGDP stabilizes to trend. The worst that could happen is Indonesia, c. 1997. (which had far more monetary stimulus than necessary).

Kenneth Duda writes:

I've heard Britonomist's objection to MM from several friends. People don't see a transmission mechanism for monetary policy at the ZLB.

My answer (for how an NGDP-targeting Fed gets traction at the ZLB) has always been: "expectations". If people expect the Fed will do *whatever it takes* to put NGDP on a 5% trend, including making up for past shortfalls, today and indefinitely into the future, then people with money will spend more at the margin, even at the ZLB, either investing more in capacity expansion in anticipation of higher future spending of others, or spending more on consumption goods out of fear of future inflation. Either way, NGDP goes up, and the Fed hits its NGDP level target.

I don't really understand Scott's answer. I think I understand Nick and Krugman's tiny models, but it's not easy for me to see how either model maps to the real world.

-Ken

Michael Byrnes writes:

As Nick Rowe often points out, people and firms experience monetary policy as the relative ease (or difficulty) with which they can sell stuff (including their labor) for money or buy stuff for money.

Recession is when it becomes easier to buy stuff for money and harder to sell stuff for money, because of a perceived change in the abundance of money.

Businesses see sales falling and also find that hiring is easier should they wish to do so. Individuals worry about losing their jobs and this drives them to hold more cash (cut back on purchases).

ThomasH writes:

Concerning the cause of the recession:

In late 2007 and early 2008, the Fed stopped increasing the base money supply.

OK that's a step toward understanding your critique, but more specifically what policy measure done or not done led to the divergence between the actual levels of monetary base and the trend level you think were appropriate?

Concerning what the Fed should have done to correct its error in 2008-09 (and later?):

With the right policy (5% NGDPLT) no QE would have been needed.

This seem to say that with the correct policy there would have been no recession to deal with but that was not my question. What should the Fed have done in 2008-09 (and later?) to return NGDP to it's pre-2008 trend (if I understand the policy objective correctly)

You do offer this hint:

The Fed should buy Treasury securities, that's all.

But which securities? Is not buying anything but ST Treasury securities called "QE?"


Kenneth Duda writes:

ThomasH:

> What should the Fed have done in 2008-09 (and later?) to return NGDP to
> it's pre-2008 trend (if I understand the policy objective correctly)

I think Scott would say: to return NGDP to its pre-2008 trend, the Fed should start by telling everyone that that's what it wants, that it will do whatever it takes to get there. The Fed has not done that. Instead, it said that it wanted to stimulate the economy, but would never risk inflation running much above 2% for very long (the "credibility of the nominal anchor"). Given that framing, it's no wonder that people just sat on the money created through QE. They knew it would all be sucked right back up again at the first whiff of inflation, that the base expansion was all temporary, intended to serve god knows what purpose.

> You do offer this hint:
>
> > The Fed should buy Treasury securities, that's all.
>
> But which securities?

I think Scott would say: it doesn't matter which securities. What matters is that the Fed expands the monetary base to whatever level is needed to hit its level-path target for NGDP. And if the Fed was clear that that was its goal, then the Fed would not need to expand the monetary base nearly as much as it had to in 2009-2014, when no one was sure what its goal was.

-Ken

Kenneth Duda
Menlo Park, CA

M.R. writes:

For those of us inclined to think the financial crisis (not tight money) caused the great recession, what's the best argument that we're wrong? It seems to me that market monetarists bear the burden of persuasion on this question, but I haven't seen a satisfying answer (which isn't to say that one hasn't been given; I admit I haven't read every market monetarist blog post).

Scott Sumner writes:

Britonomist, You asked:

"But why?"

Because they were worried about inflation.

If there is an adverse shock which increases the demand for base money, then you increase the supply by as much as demand increased (technically by as much as the Cambridge K increases.) This keeps NGDP growth steady.

Ken, Keep in mind that Britonomist was not asking about the zero bound problem, he was asking about 2007-08, when we were not at the zero bound. You are right that expectations play a much bigger role at the zero bound.

Thomas, You asked:

"OK that's a step toward understanding your critique, but more specifically what policy measure done or not done led to the divergence between the actual levels of monetary base and the trend level you think were appropriate?"

The Fed should have done enough open market purchases to keep the base growing by enough to keep NGDP growing by 5%.

Regarding what they should have done in 2008-09, after the mistake was made, my answer is adopt a 5% NGDPLT target path, from the beginning of 2008. In my view that would have been enough; no QE would have been needed. The very fast NGDP growth that would have been expected would have prevented nominal interest rates from staying at zero. No liquidity trap.

If they do need to inject more base money, then do simple OMPs with Treasury securities.

M.R., Google my papers entitled "The Real Problem was Nominal," which are devoted to showing that tight money caused the Great Recession.


Britonomist writes:

"Because they were worried about inflation."

Inflation reached a high level in 2005 as well.

In general there's something very alarming about the idea that a small reduction in the growth of the monetary base can have such devastating effects on the nominal and real economy, I can't see any simple model that would produce such a severe contraction (of course if you actually consider financial acceleration, then it begins to make more sense, but Sumner tends to dismiss finance and banking in models). To me that screams unsustainable system, if even a minor attempt to stop inflation by slowing base money growth causes such a huge contraction, then something is very very wrong structurally that begs explanation.

This coupled with the fact that I don't see how base money adjustment can offset a credit shock means I still don't see this explanation as very convincing at all.

Incidentally, here's an interesting paper I found:

http://english.mnb.hu/Root/Dokumentumtar/ENMNB/Kiadvanyok/mnben_mnbszemle/mnben_szemle_cikkei/bulletin_2007june_komaromi.pdf

"The article explains the direction of the actual mechanism and argues the point that, contrary to the view still widely held in academic circles, a great deal of the factors affecting the monetary base are exogenous for the central bank. Accordingly, the growth rate of M0 (monetary base or base money) carries no direct information on either the intentions of the central bank or the outlook for inflation."

Jose Romeu Robazzi writes:

@Britonomis, Prof. Sumner
Leverage in the financial system ("large term mismatch financing") may have a multiplying effect in what would otherwise be a mild decline in real output and rise in inflation. IT forces the monetary authority do reduce further output in a system that has seen already a reduction in output. That may cause increases in loan delinquency. If leverage is too big, banks will fail, increasing risk perception. Risk averse agents (all of them) will scramble for cash (safe assets). Velocity of money goes down, PQ goes down. But please bear in mind that a "well financed (no large term financing mismatch)" economy might have only minor bumps ... This is the hypothesis I thinks has a good explanatory power of 2008-2015.

Michael Byrnes writes:

Britonomist wrote:

"In general there's something very alarming about the idea that a small reduction in the growth of the monetary base can have such devastating effects on the nominal and real economy, I can't see any simple model that would produce such a severe contraction (of course if you actually consider financial acceleration, then it begins to make more sense, but Sumner tends to dismiss finance and banking in models)."

It is not the "small reduction in the growth of the monetary base" that matters - it is what that reduction implies about the future path of policy.

It's like steering a ship. Turn the ship imperceptibly, infinitesimally to port, and instead of reaching your destination you will travel in circles. Of course you won't, because someone would eventually notice was happening and correct course. But what if no one does?

Comments for this entry have been closed
Return to top