I Don’t Play Favorites When It Comes to Market Monetarism
I finally met Scott Sumner at the APEE conference this week. We shook hands, exchanged pleasantries, and basically were civil human beings. But that will not stop me from making this post. The fate of the US economy hangs in the balance. We must not let personal associations cloud our duty.
In a recent EconLog post Scott wrote:
This post is about the way I think about extreme outliers. It’s very unscientific, but I hope the comment section will help me to better understand this issue.
Suppose you have two variables, X and Y, which are (supposedly) positively correlated. But the very highest value of X is associated with very the lowest value of Y. Or assume the two variables are supposed to be negatively correlated, but the highest value of X corresponds to the highest value of Y. Should that make us suspicious of the alleged relationship?
…
Suppose you believed that direct democracy led to bad political outcomes, because philosopher kings were much better than mob rule. In that case, how likely it is that the one country with by far the most direct democracy in the entire world, would also be arguably the best governed in Europe, and perhaps the world? Possible, but how likely?Suppose you noticed that America scores higher on happiness rankings than does Europe, on average. So you developed a hypothesis that social welfare states are less happy. How likely would be that a country which by some measures has the world’s most generous social insurance system, is also the world’s happiest country?
And while we are on the subject, suppose you thought deregulation and privatization made people unhappy. How likely would it be for the world’s most free market economy (excluding level of taxation and government spending) to also be the happiest?
Suppose you thought that the East Asia tiger economies were successful because they rejected the neoliberal agenda coming out of Washington, and instead had state directed development strategies. If that were true, how likely is it that the two very richest East Asian economies would also be number one and two in the world in the Heritage Ranking of Economic Freedom?
Suppose you believed that monetary policy was ineffective at boosting NGDP at the zero bound. In that case, how likely is it that the fastest 4 month stretch of NGDP growth in American history would occur during a period of near-zero interest rates, right after a easily identifiable monetary shock (March-July, 1933).
And while we are at it, suppose you believed that the credit channel explains why growth is slow during and after a banking crisis. How likely is it that the fastest stretch of NGDP growth would occur during a period right after America’s worst banking crisis, and during a period when 1000s of banks were still closed down? Again, not just growth during the financial crisis, but perhaps the fastest NGDP growth ever, during arguably the worst banking crisis ever.
Suppose you thought that inflation was caused by bottlenecks in the economy, and deflation was caused by slack. How likely is it that the price level (WPI) would rise by 20% during a period of 25% unemployment (1933-34)?
And speaking of the zero bound, just how likely is it that the biggest two day stock rally in US history would (just randomly) occur immediately after Hoover announced a proposal to allow the Fed to print more money, for each ounce of gold backing.
Suppose you thought that Mexican-Americans had a propensity to rape and murder. (Hmm, where have we heard that theory?) How likely is it that America’s most Mexican major city (of the top fifty) would also have the lowest murder rate, and perhaps the lowest violent crime rate?
Here’s how I look at it, and I want you to tell me why I’m wrong. If you have only a few observations, then extreme outliers are no big deal—but your study is also not very reliable. If your study includes a large number of observations, the odds of the most extreme value of X and Y being correlated in the opposite direction from the actual relationship seems very low. Am I too suspicious of extreme outliers? What do you think?
You get Scott’s point, right? He’s (of course) being a bit tongue-in-cheek, and saying that these theories are all goofy. He thinks it’s incredibly awkward for the people who hold these theories.
I have one more to add to the list:
Suppose you thought that the Fed’s tight money policy caused the Great Recession. How likely is it that the monetary base grew far, far more from mid-2008 to mid-2010 than in any other two-year period in US history?
P.S. Of course Scott will say, “In this particular case, the growth in the monetary base doesn’t correlate with loose money, even though it normally does. It’s a poor indicator.” Right, but that misses the whole point of Scott’s post. Any defender of any of the theories Scott is mocking can explain those “outliers” on other circumstances too. They sleep soundly at night, thinking themselves good scientists, just like Scott sleeps soundly at night, thinking “tight money” describes 2008-2010.
How many checkmates against market monetarism does a checkmater check if a checkmater could checkmate?
I remember that Steve Hanke said money was tight during that period due to Basel III increasing capital requirements which resulted in banks reducing risk assets (because stock price was less than book making it unattractive to increase the capital side of the equation). So bank money was very tight despite a huge increase in base money. He says bank money is way more important than base because it’s much larger.
I realize that’s just explaining away contradictions but I find it ironic that I use that same explanation to explain the tack of an inflation response – the extra money is tied down on the sidelines.
Great post Dr Murphy. I just find it odd how mainstream economists use one variable and then switch to describe another so as to argue their point when they can’t understand why one of them doesn’t produce the intended consequence they try to describe.
For example, I read in one of Sumner’s blog posts where he wanted to target NGDP growth by looking at Real Rates by proceeding to describe how 10-year TIPS were at 4% in 2008. I think there he was trying to suggest that the monetary base doesn’t correlate with easy money. Easy money to him would have been lower Real Rates.
What surprises me is the narrow definition that mainstream economists use to define inflation. Do they not know the monetary base is inflation? Even if the Fed unwinds it’s positions by selling treasuries, that will still cause deflationary pressures and an “inevitable” downturn will occur. There is a reason why Austrians call this moment in time “inevitable.”
The continuous desire for economists to tout inflationary measures to boost macroeconomic principles misses the micro side and the resulting malinvestments that occur. It is not true that fundamental analysts or mathematical forecasters will see what is going on, even if financial statements are vetted to the T. Just yesterday I noticed certain market forecasters suggested the market is more over bought than in 2009, while another suggested the market is more oversold than in 2008.
Capital consumption is not prevalent to those receiving false market signals of distorted future expectations. It is for this very reason that excessive debt issuance and other forms of financial engineering seem to not become a problem for investors until interest rates start to rise.
If you listen to Donald Trump and Peter Schiff it’s already scrunted. Forgot about the defibrillator, reach for the bucket and sponges.
Thrn again, both of those guys are selling something… so they might be tempted to over egg the pudding. I think Americans have huge capacity to reinvent themselves… once they have put an end to socialism.
I don’t get what he is saying here, are there cases where people believe that price inflation is completely unaffected by money printing? I mean there is always the possibility that more than one factor is involved, which is kind of why someone invented statistical regression in the first place.
I find Scott’s examples carefully contrived … he supports Democracy in one breath while citing the benefits of Singapore in another. Hate to have to mention this Scott but the opposition in Singapore has never won an election. Is that the type of Democracy that Scott supports? The Democracy in Switzerland decided to disapprove of Muslim cultural influences and approve of strict controls on citizenship, would Scott agree with the Swiss , knowing the politically correct position right now? I guess not.
Well you heard that theory nowhere Scott, except perhaps your own fevered imagination, or from a dishonest journalist paid to deliver an adgenda. Oh wait, you are trying to malign Trump, am I right? But Trump was talking about illegal immigrants, not people settled in for several generations.
If Scott is talking about Laredo, Webb County, Texas (96% Hispanic), then I should point out that Trump did better than both Rubio and Cruz in that city (overall in Texas as a whole Cruze did much better, which might be another reason why someone invented statistics; but Cruz also talked tough on illigal immigration, even though he didn’t mean any of it). The murder rate is low in Laredo simply because it is low in all small towns, and pretty much universally higher in big cities… maybe that’s the point Scott was trying to make, but maybe not.
Possibly Scott is talking about L.A. and we haven’t seen the results of the primaries over there yet, but my point is that people who have settled in legally have no particular reason to be enthusiastic about others coming to compete with them just because those others are coming from the same place. Some people think in terms of the tribalism that the media proscribe for them, but quite a lot think from their own point of view.
Anyone who searches hard enough can find counterexamples to any economic theory.
Point is: how many can you find, is there a trend, and did you have to cheat a bit to find them? I really dislike the examples of convenience that are used once and thrown away. Doubly so when if you use the same examples in a slightly different way you get a completely different result.
“The murder rate is low in Laredo simply because it is low in all small towns, and pretty much universally higher in big cities… ”
Compare Wilmington Delaware’s murder rate to NYC’s!
No one is insane enough to commit a massive terrorist attack in Delaware.
Scott’s just isolating single factors as predictive to certain outcomes. Any school can data mine their theory into relevance.
Question: has there ever been a recession during which NGDP growth was flat? if there has been one, I would be much more comfortable teasing Scott Sumner…
Not sure I understand your question. Do you mean, has there ever been a recession where the rate of NGDP growth did not decrease?
During the recession of 1973-75, average annualized quarterly growth rate of NGDP was 7.6%-in the previous four quarters it was 11.1% which is less but…
The standard deviation of the previous four quarters was about 3.6 points, so that’s a less than one standard deviation decrease. But wait, it gets better: Some Market Monetarists think Aggregate Demand is better measured by Final Sales to Domestic Purchasers. Those same growth rates? 9.2% during the recession and 8.9% in the previous four quarters. In other words, if FSDP is a better measure of Aggregate Demand, then the growth rate of Aggregate demand *increased* during that recession. But even if you *don’t* think it’s a better measure, the decrease to which you want to attribute a recession is so small that whether a decrease in the rate of change occurred *at all* is entirely dependent upon your choice of AD measure.
Well when recessions are defined in terms of NGDP…
To expand on what M_F said, the trouble with your question is that a recession is defined as two (or three, or whatever) consecutive quarters of negative NGDP. I suppose you could look at industrial figures or unemployment rates over some period of time and say that economic conditions were bad during some period of time when NGDP was flat, but you’d get some pushback.
If you think those other factors are more important than NGDP, you could say, I suppose, that Market Monetarists operate within a tautological framework, but again I think you’d be in the minority.
During 1970’s stagflation (1973-75) RGDP fell but NGDP increased at a brisk rate (inflation was high).
I think Sumner would say RGDP fell due to supply-side factors, but this does seem a bit awkward for his views.
Levi- Two quarters of negative RGDP growth is a folk definition. NBER denies this is how they determine when recessions begin.
On the other hand, it’s not clear how they *do* decide, so the press generally goes with the folk definition.
Guys, by recession I mean the Austrian definition: the cluster of entrepreneurial errors, firms going bankrupt at the same time, the cluster of layoffs and unemployment. Think 2008 and the Great depression.
(btw, it seems a bit silly to define a recession as falling NGDP, as this would assume what is under discussion, i.e. that a central bank alone can avoid recessions by printing money.)
We need to be able to agree on what historical events have been recessions if your question is to be answered.
Only one recession in the post-WWII era in the US seems to me to obviously fail to meet your definition, most because it’s dubious whether it constituted a recession at all-that of 1945. I’m not going to guess as to what events you would accept as potential examples, you’re going to have to tell me.
Hi Andrew, well, I would not define a 2% increase in price inflation for a couple of years as a recession. this is how NGDP stabilization is supposed to work in the presence of a negative supply shock.
I can’t be more precise, but for the US I’d accept more than 7% unemployment as a recession. I probably should also consider a big drop in RGDP as a recession.
In the meantime I found this list (https://en.wikipedia.org/wiki/List_of_recessions_in_the_United_States), I looked at the most recent recessions, and it seems that they all either had a big NGDP decline, or were very short (like the 2001 dot-com bubble).
“I would not define a 2% increase in price inflation for a couple of years as a recession. this is how NGDP stabilization is supposed to work in the presence of a negative supply shock. ”
So what you’re saying is, if the Fed were successfully targeting NGDP, and RGDP falls, you would not consider that a recession? This strikes me as begging the question.
“I can’t be more precise, but for the US I’d accept more than 7% unemployment as a recession. I probably should also consider a big drop in RGDP as a recession.”
In this case, the 1973-75 recession in the example I gave above, should qualify. But you’re misreading Wikipedia’s chart. The peak to trough GDP decline is *RGDP* The only recession since the 1950’s in which NGDP growth was actually *negative* was the most recent one.
Maurizio,
Speaking for myself, I do not separate the “clusters of errors” and spending (NGDP), at least in the short run. Clusters of errors are manifested in part by clusters of money losses. The other part is of course clusters of errors in capital coordination.
Now profits are of course the difference between money costs and money revenues. If the assumption is made then that “NGDP keeps growing at 5%”, this tends to bring about a 5% addition to aggregate profits. For the sake of argument, imagine every dollar of spending today is invested in capital and labor. Imagine that spending is 1000 units of money, where 1000 represents however many trillions of dollars that make up aggregate spending. If we assume there is no depreciation, that is, all capital is used up in the first year, then we know that money costs will be 1000 this year. Then imagine time goes on, and NGDP grows. If NGDP grows by 5%, then aggregate revenues will become 1050. We just calculated aggregate costs to be 1000, so aggregate profits will be 50, or 5%.
If the assumption is made then that NGDP keeps increasing year after year, or at least does not decrease, then we are simply assuming that aggregate profits will remain roughly positive each year, or does not decrease. So in money terms, there can be no aggregate losses. In other words, losses in dollar terms are, roughly, a function of this period’s NGDP and last period’s NGDP. As long as that difference is assumed as remaining positive, or at least non-negative, then there will be no aggregate losses in dollar terms.
This is where market monetarism ends. It is sufficient that dollar profits in the aggregate remain positive, which requires NGDP growth to remain positive, and this is assumed as preventing “demand side recessions” in the aggregate….in the short run. I say in the short run again for a good reason, more below.
So if I had to answer your followup question, if the assumption is made that NGDP growth remains positive, then, at least in the short run, there will be no “cluster of errors” manifested in money losses.
But clusters of errors will still have been made. Here is where Austrianism does not end the way market monetarism ends. While market monetarism cannot go further than whatever “ideal” conditions are present for nominal “demand”, i.e. spending, Austrian theory has something to say about the real coordination, or rather the discoordination, as it is associated to whatever divergency from unhampered calculation the monetary conditions happen to be.
Even if central banks print whatever quantities of money necessary to force NGDP to grow at 5%, and again this is my own view, even if there would be no aggregate losses in money terms, the capital coordination errors will multiply. The capital coordination errors will not go away, nor be attenuated after a long enough time of 5% NGDP growth. They will build. And build. At the same time, the need for ever larger doses of inflation will be needed to prevent the real corrections so as to keep NGDP growing at 5% while market forces put ever greater downward pressure on it.
At some point, objective reality will establish itself as a determinant against the effectiveness of inflation to keep the growing number of errors from being corrected by investors in a division of labor. Now if the counterfeiters are obstinate, and they continue to inflate, then what happens is that we observe a “breakdown in the monetary system.” The real discoordination goes to infinity, which requires infinite inflation to sustain, which is impossible.
What market monetarism is at root, is an article of faith that with a total absence of market forces in money production, and forced spending increases, that investors will find a way to stabilize capital coordination and essentially adapt to the absence of competitive market forces in the production of money. This is why the “EMH” is promoted so heavily.
In summary, in the short run a stable growing NGDP forced by central banks would tend not to contain the cluster of monetary losses, but in the long run it cannot force objective reality to bend to its will.
‘Suppose you thought that the Fed’s tight money policy caused the Great Recession. How likely is it that the monetary base grew far, far more from mid-2008 to mid-2010 than in any other two-year period in US history?’
isn’t one of Sumner’s main points that he sees no direct correlation correlation between tight money and money base growth?
In fact you can reword what Bob wrote as:
“Suppose you thought that increasing the monetary base can never be considered tight money. How likely is it that the monetary base grew far, far more from mid-2008 to mid-2010 than in any other two-year period in US history, and yet NGDP growth told us that we actually had tight money during that time ?”
And (from Sumner’s viewpoint) score a point against Bob’s view.
That’s the point though. Since countless other factors must be involved in the outputs being measured, it’s not meaningful to look at one factor and one outcome and declare a predictive correlation. Each event discussed will have its own unique circumstances and forward-looking externalities.
Hi Bob, First off, I am friend,not foe, to wit: I recently purchased – young economist – for my 15 yr old daughter. ( now I have to incentivize her to read it…;)
Having said that, I really feel you are barking up the wrong tree here. I have been reading Scott for yrs and I see him as a straight shootin old school midwesterner. For me-He was actually thinking out loud about whether and when outliers disprove a theory.
so when he says:
I hope the comment section will help me to better understand this issue.
I think what he means is:
I hope the comment section will help me to better understand this issue.
The idea that he wrote the post as a thinly veiled insult to his intellectual inferiors strikes me as wrong and a bit paranoid. I may be a naive, but I feel he wrote the post in good faith. The comment section certainly didn’t reflect your view that the post was meant to be snide.
Keep in mind Bob, that when you comment at money illusion he shows you full professional courtesy, i.e. longer and fuller responses then the civilian commenters get. He shows you the same respect he reserves for Nick Rowe, Sadowski, Yudkowsky et al.