Arnold Kling  

On Monetarism

Greg Mankiw Gets Technical... EconLog Book Club: For a Ne...

In a comment on this post, Scott Sumner writes,

If the Fed doubled the money supply would you suddenly change your preferences and want to hold twice as much cash in your wallet?

To which I retort: if you suddenly found yourself with twice as much cash in your wallet, would you double your spending?

When the Fed lends money to banks in the repo market, by purchasing Treasury securities that the bank agrees to repurchase a week or so later, it causes ripples in the financial system. Various institutions adjust their portfolios in response. My view is that these ripples are quite small in the grand scheme of things. Long-term credit markets (where mortgage rates are determined) may ignore the ripples. Markets for risky corporate debt may ignore the ripples.

Think of there being many, many money supplies: M0, M1, M2, M3, ... The Fed causes some ripples in M0 and M1, but to affect the economy it needs to affect M75 or somesuch. And the ripples die out by the time we get there.

Yes, I am an outlier in this belief. Krugman only argues that we are in a liquidity trap when interest rates are near zero. I think that we are in a liquidity trap any time we do not have inflation approaching double digits.

Bryan says that he thinks of the LM curve as vertical. I guess think of it as horizontal, because I think that money demand is very loosey-goosey. Sometimes you wiggle M0 and you get a response in M4. At other times, you only see a response in M3. It depends on what the institutional habits are at the time. But if you need to get a response from M75, then good luck.

I really don't like to talk about LM curves and IS curves, though. I don't like to use "secret code" that some readers are not familiar with. Moreover, I think IS-LM confuses even those of us who know what the terms mean.

The most important thing going on in financial markets right now is de-leveraging, which means that everybody wants to sell risky assets and buy U.S. Treasuries. Where do you stick that in the IS-LM framework? Treat it as an increase in liquidity preference and shift the LM curve? Treat it as a drop in animal spirits and shift the IS curve?

It's hard enough to figure out what is going on. Trying to translate what is going on into IS-LM jargon just makes it harder, at least for me.

Comments and Sharing

CATEGORIES: Macroeconomics

COMMENTS (10 to date)
Bill Woolsey writes:


Sumner is arguing that an increase in the money supply will cause increase in nominal income.

If an increase in the quantity of money simply leads to a matching increase in money demand, this will not happen.

However, as long as the demand for money rises less than the quantity of money, then spending will rise.

The smaller the increase in spending from any given increase the in the money supply, then the larger the increase in the money supply necessary to generate the needed increase in spending.

I am curious. What did Sumner say to make you think that he believes that changes in the money supply must lead to proportional changes in the nominal income?

Did he say that he is a monetarist somewhere, and that is what being a monetarist means to you?

Dezakin writes:

How can you economists know which of these macroeconomic theories models reality? It seems there are a lot of theories that are contradictory and people just pick what their preferences are? Is there more to it?

Krugman thinks x, while I think y, and Brian thinks z. Here are some anecdotes, and heres a couple of metaphors... Aren't there macroeconomic software simulations or something that actually give us some idea of what is going to happen when we pull levers on these huge machines or is it all on faith?

I mean, we can show some obvious things like trade is good most of the time, good old Ricardo comparative advantage. Or that markets usually function better than planning commissions at resource allocation (maybe? I think that but its an argument made by analogy and metaphor...)

But when talking about money supply, aggregate demand, deflationary or inflationary cycles, systemic risk, moral hazard, etcetera... How the hell are we able to make any predictions at all?

I mean it seems we can make predictions about what would happen in the extremes, like if we put the tax rate to zero and powered federal spending on the federal reserve's printing press bad things happen, and we can look at Zimbabwe for a case study. But it doesn't seem like we're able to predict anything useful.

If we could actually predict these things, we would be able to shift the public policy debate to whats desirable rather than giant guesswork experiments to solve problems we're not even sure we have.

Niccolo writes:

I don't think any monetarist would have ever suggested you would have always doubled your spending.

I always thought of increases in the money supply as not having a 100% effect. I don't think that's ever been contested.

Greg Ransom writes:

When the Fed doubles the money supply, a lot of people get that money before I ever see any of it.

As Hume and Cantillon helped everyone understand, this matters.

My demand for money will go up once it becomes obvious that I need lots more of it to do what I once did with much less. This happens after lots of other people get to use that massive influx of new money before I do.

So the answer to Scott question, "If the Fed doubled the money supply would you suddenly change your preferences and want to hold twice as much cash in your wallet?" is evident. My preferences will change after everyone else reaps the benefit of that new money, which actually won't take that much time. So the force of the word "sudden" in the question is simply relative to different people's interests. It won't be sudden enough for me, but it will be sudden enough for the guys who get to use the money first.

Arun Eamani writes:


Of general interest and not pertinent to this post, is how in the Financial past, many of the enterprising managers/suits learned Finance without proper appreciation of the
economics behind it, did this knowledge disconnect impact the decisions that were made like using time series models indiscriminately without understanding of the structural aspects?

El Presidente writes:


[I]f you suddenly found yourself with twice as much cash in your wallet, would you double your spending?

Yes. Yes, I would.

Bill Woolsey writes:

Would you double you spending?

This doesn't make any sense without a time element.

Per week? Per month? Per year?

The market process that doubles nominal income is that _all_ of the excess money is spent. The increased spending (which is more than double the current rate of spending over very short periods and less than double the current rate of spending for longer periods) raises nominal inocme.

The higher nominal income increaes the demand for money. If real income is equal to capacity, then this only causes an increase in the price level. When the price level doubls, the real supply of money is back where it started. And so, the there is no longer any excess money to spend.

Given lots of ceteris paribus assumpions, in equilibrium, total spending has doubled.

If it is true that the current price level is above equilibrium and real income and output is below capacity, then the increase spending raises real income and prices. The higher real income will raise money demand, but it not neessarily in proportion to its increase. It depends on the income elasticity of the demand for money.

George McCandless writes:

It is not very useful to talk about the effects of increases in the money supply without talking about exactly how the increase will happen. The effects of a money transfer to families, seigniorage, or monetary injections into the financial system are all different (a demonstration of this using a simple general equilibrium model can be found in my book: The ABCs of RBCs). One needs to be precise about how the money injection will enter the economy, what its direct effects will be and on which agents before any monetarist analysis makes sense. I agree that IS-LM models are not especially helpful, specifically because they ignore these issues.

dearieme writes:

Your remarks have prompted the happy thought that if I bruit it about in the pub that I have krugerrands buried in my garden, I will probably get it dug over free.

Noni Mausa writes:

I'm with Greg Ransom. The first flood of TARP money, as far as I can see, never made it past Mo. Reminds me of Dean Swift's floating country, Laputa.

We saw real productivity grow over the past generation, with only nominal. not real increases in salaries to the staff. The fiddling of the Fed never affected the lowest 2 quintiles -- except to throw a spanner in the form of interest rates in the 80s. (I remember that because I landed my first good professional job in 81, my husband had a good job too, and still we couldn't afford to buy a house.)

If Obama can manage to inject an increase in wealth and prosperity to the lowest 2 quintiles, I'll be very pleased. It'll be a novelty, after the past 30 years.


Comments for this entry have been closed
Return to top