The connection between DeLong’s lament and Friedman’s lament

Nearly 15 years ago Milton Friedman was shocked to discover that he was one of the very few economists who understood that low rates don’t mean easy money, something he had assumed was pretty obvious:

Low interest rates are generally a sign that money has been tight, as in Japan; high interest rates, that money has been easy.

.   .   .

After the U.S. experience during the Great Depression, and after inflation and rising interest rates in the 1970s and disinflation and falling interest rates in the 1980s, I thought the fallacy of identifying tight money with high interest rates and easy money with low interest rates was dead. Apparently, old fallacies never die.

More recently Brad DeLong was shocked to discover than most economists didn’t think there was an obvious need for more monetary stimulus, even when NGDP had suddenly plunged 10% below trend:

And I thought that economists had an effective consensus on macroeconomic policy. I thought everybody agreed that the important role of the government was to intervene strategically in asset markets to stabilize the growth path of nominal GDP. I thought that all of the disputes within economics were over what was the best way to accomplish this goal. I did not think that there were any economists who would look at a 10% shortfall of nominal GDP relative to its trend growth path and say that the government is being too stimulative.

I’m not going to criticize the naivete of these two outstanding economists, as I made both mistakes.  But I’m increasingly convinced that there is a connection between these two complaints.

Ask yourself why poll after poll shows that most economists do not think monetary policy is too tight.  Indeed economists were skeptical of the need for additional monetary stimulus even before QE3, which means the Fed is actually more stimulative than what the median economist would prefer.  I have to think that the only reason most economists believe the Fed has done enough is that they (wrongly) believe the Fed has already done a lot.  What other explanation could there possibly be? Surely if the fed funds target was currently 8.25%, most economists would favor lowering it to at least 8.0%.  They see the need for more AD, they just don’t want the Fed (and perhaps Congress as well) to do any more.

So that’s the connection; because economists don’t understand that money is tight they are content to let NGDP languish at low levels, without demanding additional monetary stimulus.

It wasn’t just Milton Friedman who said low rates don’t mean easy money, so did Mishkin and Bernanke.  I thought if the number one money textbook author, the number one monetary policymaker, and the number one monetary economist all emphasized that low rates don’t mean easy money, then the profession would have absorbed that message.

I was wrong.

Milton was right; apparently old fallacies never die.

PS.  There’s a new study that contains the actual transcripts of the Bretton Woods Conference.  The editors are Kurt Schuler and Andrew Rosenberg.  I look forward to reading it after the semester is over.


Tags:

 
 
 

51 Responses to “The connection between DeLong’s lament and Friedman’s lament”

  1. Gravatar of John John
    25. October 2012 at 04:49

    I can’t stand Delong personally. He never mentioned a thing about nominal GDP in 2008 or 2009 and assumed that the Fed was done as soon as interest rates hit zero. Now he’s acting like the NGDP thing was either obvious all along or his own pet idea.

  2. Gravatar of Benny Lava Benny Lava
    25. October 2012 at 04:53

    Shout it from the rooftop Scott!

    Sometimes I wonder if there is a 40 year lag in economics. During the 30s the Fed screwed up by tightening during a depression. Then 40 years later the Fed screws up the opposite way. Now 40 years later we get the inflation hounds we needed 40 years ago, not now. Seems like the economics we get today is the policy we needed 40 years ago.

  3. Gravatar of Kevin Donoghue Kevin Donoghue
    25. October 2012 at 04:55

    “I have to think that the only reason most economists believe the Fed has done enough is that they (wrongly) believe the Fed has already done a lot. What other explanation could there possibly be?”

    Some of them probably agree with Richard Koo:

    Well, actually, the Japanese experience told us that when private sector is minimizing debt or deleveraging with very low interest rates, there’s very little monetary policy can do.

    In fact, Chairman Bernanke has been saying, since the middle of last year that this is no time to cut budget deficit. The fiscal stimulus should be in place because I think he also understands that under the circumstances, there’s so little that monetary policy can do.

    But there’s a lot the fiscal policy can do to keep the U.S. economy from losing its bottom.

  4. Gravatar of Britonomist Britonomist
    25. October 2012 at 05:00

    Embedded in this post is the assumption that the Fed can easily boost NGDP, and that the reason NGDP isn’t as high as it could be is because of ‘tight money’; thus lower NGDP shows money to be tight and is ‘all the evidence you need’, but that is an entirely circular argument (money is tight because NGDP is low because money is tight because NGDP is…). You need to come up with a meaningful definition of monetary tightness INDEPENDENT of your assumed ultimate outcome, otherwise you’ll never be able to convince anyone who aren’t already monetarists.

    In the olden days monetary tightness had a simple definition: banks literally had difficulty attaining funds (literally the fed or other banks were tight with their money), hence interest rates would rise as demand for funds increased, affecting the cost of credit, reducing demand. But that is CLEARLY not the case now, or if it is you need to show that it is, because as far as I can tell banks can easily get funds cheaply if they want to expand credit, it an excess demand for money can’t be a problem facing banks. So in what meaningful sense can money said to be ‘tight’?

  5. Gravatar of RPLong RPLong
    25. October 2012 at 05:04

    Suppose in time period t, the Fed reduces the interest rate and embarks on an expansionary monetary policy regime in response to tight money, as per Friedman’s quote above.

    Further suppose that in time period t1, the Fed further reduces the interest rate in ways that are not justified by any valid monetary policy scheme, including NGDPLT. That is, suppose the Fed reduces the interest rate to a point much lower than is responsible or economically justifiable.

    In time period t, Friedman will say what he said and be correct; in time period t1 Friedman will say what he said and be totally wrong; and in time period t2 Sumner will say what Friedman said and will cease to make any sense (to me) at all.

    We can’t drop the context here. What Friedman said can only be true in all periods if it is apodictically true for all M and all I and all t. Were that the case, he never would have chosen to use the word “generally.” If the Fed lowers interest rates to an irresonsible level and induces “the worst economic crisis since the Great Depression,” why should economists believe that further monetary expansion is the proper course to take?

    Or, am I totally mistaken?

  6. Gravatar of Bob Murphy Bob Murphy
    25. October 2012 at 05:16

    The first commenter “John” nails it. Here’s a sample of what DeLong was actually saying when the crisis first struck. This is from January 2009. He first quotes Gary Becker, who had been questioning the need for a big Obama stimulus package when people weren’t expecting the recession to be worse than the 1982 one, and then DeLong explains why Becker is wrong to argue in this way:

    “The difference between now and 1982 was that back in 1982 the interest rate on Treasury bills was 13.68%-there was a lotof room for the Federal Reserve to cut interest rates and so reduce unemployment via monetary policy. Today the interest rate on Treasury bills is 0.03%-there is no room for the Federal Reserve to cut interest rates, and so monetary policy is reduced to untried “quantitative easing” experiments.

    The fact that monetary policy has shot its bolt and has no more room for action is what has driven a lot of people like me who think that monetary policy is a much better stabilization policy tool to endorse the Obama fiscal boost plan.

    The fact that Gary Becker does not know that monetary policy has shot its bolt makes me think that the state of economics at the University of Chicago is worse than I expected-but I already knew that, or rather I had thought I already knew that.”

    I mean, if I didn’t tell you guys that DeLong wrote the above, you would’ve thought it was Scott Sumner, right…? And someone remind me: What part of the Obama stimulus package involved a “strategic intervention in asset markets”?

  7. Gravatar of Steve Steve
    25. October 2012 at 06:03

    Benny Lava wrote: “Sometimes I wonder if there is a 40 year lag in economics.”

    Yes, there is. Economists get educated in their 20s, climb the ladder for four decades, and make policy during their 60s.

  8. Gravatar of johnleemk johnleemk
    25. October 2012 at 07:20

    Agree with Steve and Benny Lava. It’s ok, Evan Soltas and Yichuan Wang will sort us out in 40 years!

  9. Gravatar of Nick Rowe Nick Rowe
    25. October 2012 at 07:21

    Very good post Scott. I too was wondering if monetary policy had “shot its bolt” back around 2008/9. Because the “monetary policy=setting interest rates” idea was so deeply ingrained.

  10. Gravatar of Adam Adam
    25. October 2012 at 07:22

    As Noah Smith keeps saying, very few people, including very few economists, understand monetary policy. And he includes himself in the group that doesn’t understand.

    I’m more cynical than you, though. I happen to think that politics and self interest play a big role. Some economists think money is loose because that’s the GOP’s position (consciously or not).

    Also banks, who hold huge piles of nominally denominated fixed rate assets fear any uptick in inflation, thus any chance of looser monetary policy scares them. And, of course, who retains the services of a lot of economists? To whom does the financial press, which many economists read, sell it’s services?

  11. Gravatar of johnleemk johnleemk
    25. October 2012 at 07:27

    BTW, Scott, looks like your previously-preferred POTUS candidate Gary Johnson is equally terrible on monetary policy: http://www.bloomberg.com/news/2012-10-25/the-presidential-candidate-who-would-destroy-the-economy.html

  12. Gravatar of Ritwik Ritwik
    25. October 2012 at 07:29

    Nick

    Friedman’s argument for saying so was the Fisher effect.And so it worked through inflation/ inflationary expectations. His main reference being a period of very high inflation and a period of deflation. So far so good.

    You don’t think that’s an incomplete story when inflation is low, but not negative (or even that far from the mandate), and rates are at zero?

    I get that levels matter etc. But if we forget about NGDP for a moment, and evaluate Friedman’s quote with r = 0 and i = 1.5%, does it have the same implication as it does during Volcker/ the Great depression?

  13. Gravatar of Saturos Saturos
    25. October 2012 at 08:01

    But DeLong is being disingenuous; he wouldn’t have started looking at NGDP unless Scott Sumner had put him on to it.

    Interestingly Johnson agrees we are in the midst of a monetary collapse, and wants to fix it with a gold standard. He might succeed, too, so long as he pegs a sufficiently high price of gold to begin with.

    Ritwik, Friedman understood the income-expectations effect as well. See “Factors determining the level of interest rates”, a speech he gave in ’68. Which he began by emphasizing the distinction between money and credit.

  14. Gravatar of Bill Woolsey Bill Woolsey
    25. October 2012 at 08:42

    In the video, Johnson said that he favored metalic backing of the currency (a really bad idea.)

    He said he favored allowing currency competition. (A fine idea, but relatively innocuous.)

    On his website, he said he favors having the Fed focus solely on price stability (a bad idea, but not consistent with a deflationary depression as Barro suggests)

    He also says that the Fed should stop creating money, quantitative easing, and interfering in the freem market. (That is approximately crazy talk. Perhaps not creating any more money and freezing the monetary base? A really bad idea, but not crazy.)

    Anyway, I support Johnson and not because of his monetary policy views.

  15. Gravatar of Saturos Saturos
    25. October 2012 at 08:49

    Bill, why is freezing the base such a bad idea? Didn’t Friedman support it?

  16. Gravatar of Major_Freedom Major_Freedom
    25. October 2012 at 09:00

    Nearly 15 years ago Milton Friedman was shocked to discover that he was one of the very few economists who understood that low rates don’t mean easy money, something he had assumed was pretty obvious:

    “Low interest rates are “generally” a sign that money has been tight, as in Japan; high interest rates, that money has been easy.

    . . .

    After the U.S. experience during the Great Depression, and after inflation and rising interest rates in the 1970s and disinflation and falling interest rates in the 1980s, I thought the fallacy of identifying tight money with high interest rates and easy money with low interest rates was dead. Apparently, old fallacies never die.”

    Sigh, this fallacy of Friedman’s never seems to die with monetarists. The escape hatch Friedman used to deflect criticism of his statements is using the word “generally”. Folks, this means that declining interest rates is not always a sign that money “has been” tight. The word “generally” means that the contrary situation is possible. Sumner’s citing of Friedman, to make a case about the recent recession, is not as strong as it at first glance appears to be.

    Moreover, and perhaps more importantly, none of the empirical examples Friedman cited are even sufficient to confirming his thesis. For one could legitimately argue that interest rates during the 1970s would have been even higher had the Fed inflated the money supply at lower rates, and that interest rates during the 1980s would have been even lower had the Fed inflated the money supply at higher rates.

    In order to take out from the 1970s and 1980s data the thesis of Friedman, one must have already decided the thesis is correct before even looking at the data! The data alone does not “prove” either Friedman’s thesis or my thesis.

    Ironically, Friedman committed an “old fallacy that never dies” himself, namely, that of confusing correlation for causation.

  17. Gravatar of Alex Godofsky Alex Godofsky
    25. October 2012 at 09:05

    @John and others:

    I can’t stand Delong personally. He never mentioned a thing about nominal GDP in 2008 or 2009 and assumed that the Fed was done as soon as interest rates hit zero. Now he’s acting like the NGDP thing was either obvious all along or his own pet idea.

    DeLong talking about how obviously correct NGDPLT is >> Scott winning some extra Internet Points. I suspect Scott would agree.

  18. Gravatar of RebelEconomist RebelEconomist
    25. October 2012 at 09:17

    I agree with MF that the Friedman quote gets abused, but for me, the key words are “has been”. Friedman was not (as far as I know) saying that high or low interest rates mean that money is or is not PRESENTLY tight.

    DeLong and Scott call to mind a poorly educated jockey, whose only answer to a horse not winning is to whip it harder. It might work in the very short term, perhaps at the cost of longer term damage, and there might be better solutions like train the horse differently, feed it better or even just allow it to rest.

  19. Gravatar of Ritwik Ritwik
    25. October 2012 at 09:24

    Saturos

    I’ve read that speech and remember it well. The central bedrock of that speech was that the real rate stays constant, over the long run, and thus the Fisher effect and hence inflation determines nominal rates. Again, in the long run.

    The ‘income expectations’ that you talk about was not the way Scott or MaMos think about it, it was Pigou & real balances. And the first effect he talked about was liquidity and loanable funds – the usual monetary easing leads to low rates channel. Friedman’s first and second effects interact on the real rate and roughly cancel each other out, leaving his third effect to do the job over the long run.

    It’s a good speech, describing the dynamics of the short rate by assuming the long run super-neutrality of money.

    But you know what applying Friedman’s long run model to the short run dynamics of interest rates does right? It leads to Kocherlakota type errors, where he believes that to increase inflation the Fed should raise nominal rates.

    We’re talking about a state where the short real rate was positive, then dropped and became negative, and looks like it will stay there for a while. This was not the case in the Great depression, where the deflation kept real rates highly +ve for long. This was not a case that Friedman was considering when he wrote about the Fisher effect and the monetarist paradox.

    Nick explicitly argues elsewhere that when monetary policy is too tight, even REAL rates go downwards. This was not a case Friedman made. In that speech or elsewhere.

  20. Gravatar of Nick Rowe Nick Rowe
    25. October 2012 at 09:35

    Ritwik and Saturos: I’m not aware of that Friedman speech, or I’ve forgotten it. But the idea that real interest rates can go down if monetary policy is too tight seems key to me. I think I got that idea from Scott, though Scott maybe doesn’t quite say it that way.

  21. Gravatar of dtoh dtoh
    25. October 2012 at 10:50

    As I have commented before, I think the correct way to understand monetary policy is in terms of the relationship between the price of financial assets and spending on real goods and services (both consumption and investment). If financial asset prices (as measured as real risk adjusted expected returns) rise, then economic players will exchange financial assets for real goods and services thus increasing AD.

    Fed OMPs have two effects. First they raise the price of financial assets. Second, through expectations, they cause a shift in the curve… at the same price, economic players will exchange more financial assets for more goods and services.

    This is a much simpler and more accurate model of how monetary policy works than any kind of a quantity/velocity model, and if it was presented this way it would reduce the confusion on the part of mainstream economists.

    IMHO, MV is determined by the volume of spending on real goods and services not the other way around. For the most part if there is excess money (more than transactional needs) in the system, then economic players exchange the money for financial assets. This increases the price of financial assets causing an increase in AD to a point where transactional needs increase to absorb the excess money. Excess money does not directly increase AD….it only works indirectly through the financial asset price channel.

  22. Gravatar of Major_Freedom Major_Freedom
    25. October 2012 at 12:51

    RebelEconomist:

    I agree with MF that the Friedman quote gets abused, but for me, the key words are “has been”. Friedman was not (as far as I know) saying that high or low interest rates mean that money is or is not PRESENTLY tight.

    Not that I want to get TOO technical on this, but isn’t “presently” still a historical timeframe? When I say “Bernanke is inflating the money supply”, what I am really saying is that Bernanke just inflated the money in some very recent past.”

    The question then is what the heck “has been tight” really means. How far back do we go? One day? One month? A year?

    Are rates low now because of what monetary policy was like 10 or 20 years ago?

    This is yet another ambiguity that is often exploited and abused in order to advance a particular argument that uses it as a premise.

    So not only is “generally” abused, but so is “has been”.

    I wonder if Sumner believes low interest rates “always” means money “is” tight, and whether he is just communicating that via Friedman’s “generally” and “has been”.

    —————–

    The reason why Friedman and Sumner are convinced that low interest rates means money has been tight, is that they are first convinced of the theory (which is what actually has to be addressed, but so far is not), and they believe, falsely, that their theory is confirmed by history, especially the 1970s and 1980s. But the history of inflation is one that has followed a particular pattern. The Fed inflates, and that lowers interest rates AT THAT TIME, and then, after it realizes it is going to overshoot its target, it reduces inflation, which increases interest rates AT THAT TIME. Then, the same thing happens again and again.

    Drawing these events on a chart, it definitely LOOKS like history is telling us “low interest rates signals money HAS BEEN tight.”

    But this theory would be empirically falsified in cases of continually accelerating monetary inflation. Then, we would see the inflation continually pushing interest rates down IN EACH “PRESENT” TIME (which may or may not be absolutely rising interest rates), and yet monetary inflation (supply and prices) are rising.

    The “generally” is not a theoretically sound argument. It is a rather crude noticing of history.

  23. Gravatar of Ritwik Ritwik
    25. October 2012 at 15:32

    Nick

    This is the Friedman paper. http://ces.univ-paris1.fr/membre/Giraud/english/Interest-rate-Friedman.pdf

  24. Gravatar of ssumner ssumner
    25. October 2012 at 17:38

    Kevin, Koo must know very little about how the BOJ operates. The BOJ got the stable price level they wanted. For almost 20 years the CPI in Japan was virtually unchanged, just as they wanted. Then the government forced a 1% inflation target down their throat (just a few months ago). The BOJ seems to feel that opens the door to hyperinflation. They tightened money in 2000 and 2006—does that seem like a central bank that was valiantly trying to debase its currency, but just couldn’t get the hang of it? Do you think they are incapable of devaluing the yen? Why did they reduce the monetary base by 20% in 2006? Why did they raise interest rates in 2000? Why did they again raise interest rates in 2006?

    Britnomist, No, I usually use expected NGDP growth, which is different from actual. If the Fed pegs the price of a NGDP futures contract at a 5% premium, or if the internal Fed forecast is for 5% NGDP growth, then policy is neutral. More than 5% is easy money and vice versa. That assumes nothing about actual NGDP.

    RPLong, He didn’t say money is tight, he said it has been tight

    Nick, You quickly realized what was going on–I wish I could say the same about the rest of the profession. At least NY Fed chief Dudley recently admitted that Fed policy had been too tight in recent years—we are making progress.

    Johnleemk, What do you mean “previously preferred”?

    dtoh, You said;

    “MV is determined by the volume of spending on real goods and services not the other way around.”

    So the big European inflation of 1500 to 1650, which was supposedly caused by the huge inflow of gold and silver from Latin America, was actually the CAUSE of Columbus discovering America in 1492?

  25. Gravatar of Saturos Saturos
    25. October 2012 at 18:13

    Scott, but surely you can’t blame BOJ for most of Japan’s sclerosis? The Japanese government has bigger responsibilities than changing the inflation target.

    Bob, wan’t QE1 such an intervention?

  26. Gravatar of Benny Lava Benny Lava
    25. October 2012 at 18:52

    Steve and johnleemk,

    Great, so 40 years from now we will have NGPD targeting which will be totally ineffective against the bitcoin virus that destroys all money.

  27. Gravatar of Bogdan Bogdan
    25. October 2012 at 19:39

    Brad deLong might be a good economist, I don’t know, but as an historian – at least in this bizzare

    article http://www.project-syndicate.org/commentary/seventy-years-after-the-battle-of-stalingrad-by-j–bradford-delong

    he fails with a big prize!

    “Our debt to Stalingrad”…with no mention of the monstroasities of the communist regime instituted by Lenin in 1918 and consolidated in the 1930s through a permanent war againt the people, banal mass executions of “classes”, ethnicities or just individuals, opression, gulag, genocide against unruly tribes and peoples, managed inter-ethnic violence, occasional politically induced famines and so on, monstrosities at least equal in substance to those of the Third Reich; “our debt to Stalingrad” with no mention of the Soviet-Nazi collaboration through the Robbentrip-Molotov Pact, which started WW II in Europe by way of a partion of all the countries situated between Germany and the Soviet Union at the time, in what todays’s World Bank jargon calls “Central and Eastern Europe”; “our debt to Stalingrad” with no mention of the huge and paranoid purges in the Red Army leadership made by Stalin in the 1930s, which made it so ineffectual in combat and were responsbile – along with Stalin’s arrogance of being a genius strategist, much like Hitler – for the Wehrmacht’s rapid advance and the butchering of millions of Russian soldiers, poorly armed, poorly led and terrorised by the Army’s secret police, sent to fight as “meat for the cannons”; and, finally, “our debt to Stalingrad” with no mention that in the Stalingrad siege, around 14 000 Russians/Soviets, in fact, were shot in the back by Soviet Communist Party propaganda commissars, “their own”, because they seemed “suspicious” or “demoralised” and so on, and no mention of the fact that the policy of no tactical retreat – on both the Russian and the German side (you see here the same thinking, the same inhumane totalitarian fanaticism) uselessly increased the death toll to such a considerably number, both among the civilians and among the military…

    So, yeah, instead of remembering the battle of Stalingrad as the bloody battle between the two very similar, but concurrent, most terrifying and bloody ideologies and political regimes of the 20th century, let’s celebrate it Putin-style, as the begining of the “liberation of Europe”, although the war started with a very significant, forgotten, but real joint German-Russian agression, sealed through a secret alliance against the League of Nation’s international law, over countries in eastern Europe….Attaboy!

  28. Gravatar of dtoh dtoh
    25. October 2012 at 23:02

    Scott,
    You ask, “So the big European inflation of 1500 to 1650, which was supposedly caused by the huge inflow of gold and silver from Latin America, was actually the CAUSE of Columbus discovering America in 1492?”

    You’re the expert on monetary history, but I would surmise that the inflow of gold and silver (financial assets) caused a shift in the indifference curve between holding financial assets and spending on real goods and service creating excess demand which caused the inflation. (Monetary policy and external events can both cause movement along the indifference curve and a shift of the curve.)

    It’s important to keep in mind that in some circumstances (16th century Europe and at the ZLB) money can become dual purpose…both a medium of exchange and a store of value and to distinguish between the two uses.

  29. Gravatar of Ritwik Ritwik
    25. October 2012 at 23:18

    Scott

    You asked dtoh, “So the big European inflation of 1500 to 1650, which was supposedly caused by the huge inflow of gold and silver from Latin America, was actually the CAUSE of Columbus discovering America in 1492?”

    Actually, Columbus was a market monetarist.

    He looked into the future and saw a big inflation and rise in prices and incomes coming. So he responded as true market monetarist citizens do, he started spending. We needn’t bother about what he spent on. History says that one of those expenditures was the trip to America. Doesn’t really matter. What matters from a monetary perspective is that he foresaw the coming explosion in nominal incomes and acted rationally on the basis of that foresight.

    Just as slow recoveries cause financial crises, the price revolution caused the discovery of America.

  30. Gravatar of dtoh dtoh
    25. October 2012 at 23:21

    Saturos,
    Scott, but surely you can’t blame BOJ for most of Japan’s sclerosis? The Japanese government has bigger responsibilities than changing the inflation target.

    Good question. I think it would be very difficult to determine “most” or “bigger”. I think “a lot” would be a better choice description in both cases.

  31. Gravatar of Saturos Saturos
    26. October 2012 at 00:51

    Ritwik, lol, that was actually pretty good.

    dtoh, my point is simply that there’s only so much stagnation that a drop in spending can cause. They have maintained price stability, after all.

  32. Gravatar of dtoh dtoh
    26. October 2012 at 02:56

    Saturos,

    I don’t disagree that other factors have also strangled growth in Japan. (So much so that were I to expound on the point, MF would appear a paragon of brevity.)

    That said however, Japanese monetary policy has also been a disaster. Japan does not need price stability. What they need is a good dose of inflation to reduce real expected returns on financial assets in order to goose up spending on real goods and services.

  33. Gravatar of Bill Woolsey Bill Woolsey
    26. October 2012 at 02:56

    Saturos:

    If the demand for base money changes, the result is undesirable shifts in spending on output.

    I don’t believe that base money velocity is constant, much less remaining on a constant growth path.

    And yes, Friedman did once call for freezing base money and allowing private issue of redeemable currency.

    If favor the private issue of redeemable currency, but don’t favor freezing the monetary base. I think the quantity of and yield on the monetary base should vary to keep expected nominal GDP on a slow, steady growth path.

  34. Gravatar of Bill Woolsey Bill Woolsey
    26. October 2012 at 03:29

    Ritwik:

    Before the eighties, Friedman identified “loose” and “tight” money with the M2 measure of the money stock.

    If M2 fell and remained below its past growth path, that was tight money. Interest rates have nothing to do with it.

    When the “tight money” (decrease in M2) causes a depression, the demand for loans fall, and so interest rates fall.

    Does that mean that money is no longer “tight?” No, M2 is still far below the growth path consistent with spending being at trend and output being back at full employment. If M2 did recover, spending and output would recover. And so would loan demand and interest rates.

    The low interest rates occured because of the decrease in M2.

    While it is also true that if there is the deflation rate is high, real interest rates can be high, that isn’t the whole story.

    This gets to the “money and credit” confused issue.

  35. Gravatar of Saturos Saturos
    26. October 2012 at 03:43

    Bill, so Friedman didn’t have a good reason for believing that the private sector would automatically adjust the quantities of the broader aggregates to offset their velocities?

  36. Gravatar of ssumner ssumner
    26. October 2012 at 04:10

    Saturos, Two issues:

    1. The BOJ contributed to the problem during those years when NGDP fell.

    2. One could argue that zero NGDP trend growth makes the labor market less efficient.

    3. Having said that, weak productivity in overregulated services and declining population explain some of the poor RGDP performance.

    dtoh, I agree that monetary stimulus (more gold) can raise asset prices, and that this is one of the transmission mechanisms.

  37. Gravatar of Ritwik Ritwik
    26. October 2012 at 04:32

    Bill

    When tight money causes a depression, the supply of loanable funds also falls. The effect on interest rates in unclear.

    My pt is, let’s not rehash the same points. I am looking for clarification on a single specific point. I see one clear difference between the Market monetarist model and the, say, Hawtrey model. Though Hawtrey has a nominal theory of the business cycle and is thus looked on by MaMos as an intellectual fountainhead, he basically looks at low-er real rates as tight-er money and high-er real rates as loos-er money. That is a worldview more or less compatible with the Keynes-Wicksell way of looking at things, though it emphasizes money rather than credit or financial markets as Keynes-Wicksell are wont to do.

    Now the market monetarist model, as exemplified by say Nick Rowe’s upward sloping IS curve, is incompatible with Keynes-Wicksell. No matter how much MaMos love Hawtrey, they just don’t share the same model.

    Now, I want to understand where Friedman stood on this. Scott would have us believe Friedman showed high rates indicated loose money. Yes he did. But he was always, always talking about nominal rates. And hence inflation and the Fisher effect.

    Friedman’s model for real rates tends to be basically one of long run superneutrality. In case of a monetary ‘easing’, he recognizes the Keynesian downward push on real rates due to liquidity, then the Pigou upward push due to income expectations and in fact even says that the rate increase is likely to overshoot the initial rate fall. But when explaining the overshoot, he ditches the real rate logic of the first two effects and resorts to the third and final Fisher effect.

    In all, he leaves his view on effect of nominal income or inflation expectations on real rates unclear, with zero effect as the most likely prior. Do you disagree with that. If yes, why?

    And if you don’t disagree with that, then my question is – how fair is it to invoke Friedman’s low rates = tight money argument when REAL rates have themselves dropped and inflation expectations are roughly on track.

    Scott, and other market monetarists, should be invoking Nick Rowe’s upward sloping IS curve model. Not Friedman’s.

  38. Gravatar of RPLong RPLong
    26. October 2012 at 04:49

    Prof. Sumner –

    What I meant was that, in my example, when one moves from t to t1 to t2, then hindsight also moves accordingly. Therefore, in t, money “has been” tight in t-1; in t1, money “has been” tight in t; in t2, money “has been” tight in t1, and so on.

    I thought long and hard about this post. I realized that under a Sumnerian view of the universe, so long as interest rates are believed to be “low,” then one must (by definition) accept that money “has been” tight, because you’re looking at these things relatively. Whatever the actual interest rate, if it feels low, then in the previous time period, money must have been comparatively tight.

    That’s true, but the problem I have with that is that it shrinks what I call the “cognitive time-horizon.” If the Fed embarks on a long-run trend of progressively reducing the interest rate for decades at a time, then we can no longer say that money is “tight” or “easy.” We can’t even think in those terms anymore, because money in the previous time period will always be tighter than money in the current time period.

    I.e., the interest rate is no longer a relevant market signal, it’s just a number that always falls. Much the same as how in an NGDPLT regime, inflation is no longer a relevant signal, it’s just a number that always rises.

    So, sure, we can eliminate market data in such a way through monetary policy, but the problem is that these signals existed for a reason. Economic actors were using them. Interest is in some respects the price of time; inflation is in some respects the price of holding cash. We can obliterate their usefulness, but with what do we replace them?

    Taken seriously, isn’t NGDPLT a rather radical idea?

  39. Gravatar of mike smitka mike smitka
    26. October 2012 at 05:35

    You can readily access policy statements of the BOJ and other materials, including English translations (with a modest delay). That includes detailed minutes of a sort the Fed does not make available. Check before claiming the BOJ was happy with mild deflation (1998 to date).
    As to growth, with the total population shrinking, and the working age population declining since the mid-1990s, we can’t use the same growth target as for the US, where fertility plus immigration give us a growing working age population (though that slows for the next several years as baby boomers age). Still, with recessions in 1993, 1997-8, 2002, 2005, 2009 and 2011 the BOJ had ample reasons to try “easy” MP. If you look back, you’ll find a storm over the introduction of ZIRP (zero interest rate policy, Feb 12, 1999). Since then the call rate [cf. Fed Funds rate] has never topped 0.46%, and during most of the past 13 has been below 0.1%. Lots of experimentation with QE, too.
    Easy money? — well, deflation intervenes.

  40. Gravatar of Saturos Saturos
    26. October 2012 at 07:01

    Ritwik, no, expected NGDP collapse causes investment to fall and saving to rise (at each interest rate). Yes the IS slopes downward, but the MM perspective is that although tight money shifts the LM back, expected NGDP collapse also shifts IS way down (potentially into liquidity trap territory), so rates go unambiguously lower. (And New Keynesians ought to agree!)

    (Nick Rowe is probably going a bit too far with his upward sloping IS – that’s not necessary) Higher expected real growth raises the willingness to borrow, even in NK models.

    Now see if the data supports this perspective.

  41. Gravatar of Greg Ransom Greg Ransom
    26. October 2012 at 07:03

    Of course, Hayek understood this as well. But, then, he was a good monetary economist — and most who read some tiny bit of Hayek here or there are NOT.

    Scott writes,

    “It wasn’t just Milton Friedman who said low rates don’t mean easy money, so did Mishkin and Bernanke.”

  42. Gravatar of Greg Ransom Greg Ransom
    26. October 2012 at 07:10

    There is *no* way to challenge the simple, easily graded ‘models’ that make up the sum and substance of the understanding of ‘economics’ built into the brain wiring of the professors.

    Really, Scott, you are up against more than you know, and you need to address this deeper, underlying fact about your profession.

    You won’t make any progress until you challenge the pat formalisms of your profession which dictate understanding and determine field of vision — economists literally can’t see or think what you are talking about.

    Scott writes,

    “I thought if the number one money textbook author, the number one monetary policymaker, and the number one monetary economist all emphasized that low rates don’t mean easy money, then the profession would have absorbed that message.”

  43. Gravatar of Greg Ransom Greg Ransom
    26. October 2012 at 07:12

    Read some T. Kuhn, _The Structure of Scientific Revolutions_ to remind yourself of what you are up against in this rivalry of explanatory visions, Scott.

  44. Gravatar of Ritwik Ritwik
    26. October 2012 at 07:21

    Saturos

    If you can’t hold one curve constant while shifting the other curve, you don’t have a two curve model. So your LM-shift-pushes-IS-downwards is not really an IS/LM model.

    It’s this very thing that led Nick to draw the IS curve upward sloping. In your model, just take the point at which your second lower IS curve intersects the new LM. You should draw your original IS curve through this new point and the original IS/LM equilibrium. And that’s Nick’s model.

    Second, if you expect NGDP (and hence your own income) to fall, and that makes you want to increase your savings, you’re basically engaging not in consumption smoothing but in saving smoothing. I will be surprised if Scott or anyone else agrees with this.

  45. Gravatar of Saturos Saturos
    26. October 2012 at 07:26

    “If you can’t hold one curve constant while shifting the other curve, you don’t have a two curve model.”

    No, it’s expected NGDP collapse that pushes down IS. It’s entirely possible for LM to shift alone; but that’s just not what happened this time. Nick’s model won’t allow for loose money to actually lower rates, which it frequently does. (I had other criticisms, too, can’t remember them right now)

    If you expect to be poorer later, wouldn’t you save now? Isn’t that why we save for retirement?

  46. Gravatar of Ritwik Ritwik
    26. October 2012 at 07:57

    “If you expect to be poorer later, wouldn’t you save now? Isn’t that why we save for retirement?”

    Sure, but your saving will place downward pressure on rates, which will increase the borrowing demand. And then in next period, when the expected NGDP fall materialises, people will consume from their savings after all, that’s whatthey saved for – reducing their savings in that period and hence reducing the supply of loanable funds and putting upward pressure on rates.

    Basically, net effect on real rates is unclear. It depends on how you conceive your IS curve to look like. That’s what I was trying to show Bill. That he/Scott believes the IS curve slopes upwards, and I haven’t yet seen any evidence of Friedman believing the same.

    Of course you can get the same result with a downward sloping IS curve that shifts downwards on an LM shock, which is your model. But I don’t quite get why the expected/ current distinction is critical to your model. After all, if there’s no expected NGDP fall, there’s no LM shock in a market monetarist model. LM, the stance of monetary policy, is by definition expected NGDP.

    What kind of events in your model constitute LM shifting alone?

  47. Gravatar of Saturos Saturos
    26. October 2012 at 08:53

    Well, the data would suggest that IS shift dominates LM shift, and we get lower rates. That’s the kind of thing the MM blogs have been painstakingly documenting for years now.

    Scott doesn’t really use Nick’s model – and I use something in-between Bill Woolsey’s model (which is inbetween Scott and Nick) and conventional textbook models.

    I’m pretty sure I’ve heard Friedman explicitly endorsing expected income effects, but I can’t find the quotes. Nonetheless if you believe in the Fisher effect and standard expectations-augmented IS curves you should be able to derive the result in a new Keynesian model.

    Scott might say that the stance of monetary policy is by defintion expected NGDP; I tend to think of it as current NGDP (the aggregate flow of money). And monetary policy works by shifting current demand or supply for the stock of money. So my LM curve works like a traditional Keynesian one; except I don’t think the liquidity effect matters as much as the excess-cash balance mechanism (which is broader than what mainstream economists think of as “portfolio rebalancing”). I basically read the move to equilibrium in ISLM “horizontally” instead of “vertically” (it’s a nonsense to think about spending without considering what it actually is, the aggregate flow of money).

  48. Gravatar of Ritwik Ritwik
    26. October 2012 at 09:34

    Saturos

    A drop in the IS curve, or in MEC, is a hallmark of a Keynes-Wicksell recession. Combine that with a relatively flat LM curve and you have a full blown Keynesian model, irrespective of the shape of the IS curve.

    You are just force-fitting the market monetarist intuition on top of this by saying that an LM shock causes a fall in IS leading to lower rates. What if the IS dropped on its own? And because of the flatness of LM, real rates came down, but not by *enough*. That’s Keynes.

    Also, ‘endorsing’ income effects is not what I’m asking about here. It’s the net effect of those income effects (+liquidity effects) on real rates. That net effect is captured by an IS curve.

    New Keynesian models won’t help because they are explicitly Keynesian in the sense that the IS curve is free to drop down on its own without any LM shock.

    You can use a full blown NK model – that’s fine, but then let’s not say that the LM shock caused the IS to drop. That’s your own version of a MaMo model. And that intuition is captured better by an upwards sloping IS curve than an IS curve that fluctuates when LM fluctuates.

    How’d you describe loose monetary policy in your model? that higher than trend NGDP (positive LM shock) caused the IS curve to shift upwards, thereby causing higher real rates? Again, this is better captured through a fixed upward sloping IS curve. If your IS curve depends so much on your LM curve, it isn’t really an IS curve.

  49. Gravatar of Bill Woolsey Bill Woolsey
    27. October 2012 at 06:31

    Ritwik:

    I am a charter Market Monetarist and I have never assumed that the “IS” curve cannot shift down “on its own.”

    However, I generally think about saving shifting to the right and investment to the left.

    The IS shifts to the left and equals potential income at a lower natural interest rate.

    Nominal spending on output, on the other hand, is only impacted to the degree that the quantity of money either decreases with no matching decrease in the demand to hold money or else fails to increase to match an increase in the demand to hold money.

    Expectations of lower real income and output in the future appears to result in more saving and less investment and so a lower natural interest rate.

    Expecations of lower spending on output in the future appears to result in expecations of lower real income and output in the future. (People don’t seem to believe in market clearing.)

    It also appears that past experience of low spending on output will lead people to anticipate low spending on output in the future.

    And so, low spending on output in the past can lead to a low natural interest rate in the present.

    Market Monetarists insist that expecations of higher spending on output in the future can result in expecations of higher real income and real output in the future and so reduced saving and increased investment in the present, and so a higher natural interest rate in the present.

    An expansionary monetary policy does not require lower market interest rates in the present.

    Identifying monetary policy with changes in interest rates is both wrongheaded and potentially disasterous.

    While I suppose a depression could result in a decrease in loanable funds, there have usually been large increases in the the supply of loanable funds to highly credit worth borrowers, and the interest rates they are charged get reported.

    How much loan sharks are charging the unemployed and businesses facing bankrupcy don’t get reported much in official statistics.

  50. Gravatar of Bill Woolsey Bill Woolsey
    27. October 2012 at 06:38

    Saturos:

    I think Selgin’s argument about the demand for reserves is sound. If it is “perfect,” then a slow stable growth path for base money will generate a slow stable growth path of nominal GDP.

    I am not confident that it is perfect. And so, I favor allowing adjustments in base money to stabilize nominal GDP. If there is an increase in the demand to hold base money, the quantity should increase. If there is a decrease in the demand hold hold base money, the quantity should decrease.

    To the degree Selgin’s process works, such changes would rarely be needed. If it worked perfectly, they would never be necessary.

  51. Gravatar of ssumner ssumner
    27. October 2012 at 06:41

    Mike Smitka, Sure, there were periods of deflation. The BOJ doesn’t like deflation. They eased policy. The deflation ended as inflation rose to zero. At which point they tightened policy. Then they went back to deflation. Then more QE. That pushed inflation back up to zero. Then they tightened. Then they went back to deflation. Then they did more QE. What’s the big mystery here?

Leave a Reply