Scott Sumner Digs Deeper
by Mike Kimel
Scott Sumner Digs Deeper
Scott Sumner criticizes my most recent post in which I indicate that Keynesian theory explains growth rates during the New Deal era better than theories proposed by monetarists.
He starts by criticizing this, which I wrote in my earlier post.
Aggregate demand was very slack when FDR took office.
FDR showed up in Washington with a plan to start spending a lot of money and thus boost aggregate demand.
The immediate effect was to convince factories they’d be running down their inventories. That boosted producer prices. It had a much smaller effect on consumer prices because everyone knew the gubmint was going to buy a heck of a lot more producer goods than consumer goods. (The government did buy some consumer goods for the various programs, plus there was a spillover effect, but as the graph clearly shows, the action was on the producer side.)
After a bit of time, the public realized FDR wasn’t planning just a one-off, but rather a sustained program of purchases of industrial items. That led them to start using some of their idle capacity, which meant not just selling the fixed amount that was in inventory. The rate of price increases thus dropped.
GDP increased the fastest rate in the United States peacetime history since data has been kept. There was a big hiccup, of course, in 1937 when the government cut back on spending for a while.
Sumner’s most important point:
Prices didn’t start rising when FDR came to Washington with spending plans; they started rising when he began depreciating the dollar. Furthermore, the weekly rise in the WPI index was highly correlated with weekly increases in the dollar price of gold (i.e. currency depreciation.) And those changes (in gold prices) were caused by explicit statements and actions by FDR. Not by fiscal stimulus, which would be expected to appreciate the dollar.
OK. Using the cool graphical tool from FRED, the Federal Reserve Economic Database, I generated this graph of the series that from what I can tell seems to be Sumner’s favorite price index when discussing the period:
Figure 1.
Now, take a gander at the graph. And bear in mind, FDR was inaugurated in March 1933. But everyone knew what he was going to do, spending-wise, once he showed up. You can see the decline in prices halt and start reversing even before he took office.
Additionally, I’m not sure what Sumner means when he refers to the period when he says FDR “began depreciating the dollar.” There was a gold standard in place going back a long time. That means the value of the dollar was its price in gold. The price of gold was $20.67 an ounce for decades before FDR took office. It was $20.67 an ounce until the Gold Reserve Act of January 30, 1934, when the price of gold was changed to $35 an ounce. (To be precise, the government devalued the dollar on January 31, the day after the Act passed.)
The peak in the curve came in February 1934, days or at most weeks (the index isn’t that precise) after the Gold Reserve Act. Put another way… price inflation using Sumner’s measure peaked when the currency was devalued. That is precisely 100% the opposite of what Sumner wrote.
But there are some extenuating circumstances for Sumner.
(The next paragraph summarizes this story, from the memoirs of Jesse Jones.)
It seems that on October 22, 1933, Jones, the head of the Reconstruction Finance Corporation and Henry Morgenthau, then Farm Credit Administrator but soon to be Treasury Secretary, were told by FDR to come by on October 23 to devaluing the dollar by changing its relationship with gold. The three men – FDR, Morgenthau, and Jones, then went about raising the price of gold by fiat between then and January 31, 1934, when prices came to rest at $35 an ounce, a price where they stayed through 1971.
I assume that’s what Sumner is talking about. So let me modify Figure 1 to only show the period from January to October 1933.
Figure 2.
Now, recall, Sumner’s evidence that the Keynesian view is wrong and the monetary view is right is: “Prices didn’t start rising when FDR came to Washington with spending plans; they started rising when he began depreciating the dollar.”
And yet… the graph shows very clearly that prices started to rise when FDR came to Washington with spending plans, not at the end of October when he began depreciating the dollar. As is very evident from the graph, by that time prices had already been increasing for quite a while. Wholesale prices, by October 1, were up 17% from the beginning of the year. If you started in October of 1933, it wasn’t until December of 1936 before prices increased another 17%.
The point is, Sumner is wrong. He is very wrong about when prices started to rise. He is also very wrong about why prices started to rise. And since “when” and “why” are assumptions in his model, his model is very wrong.
Now, for completeness I’m going to tackle the other thing Sumner mentioned in his post. Sumner’s critique of me includes this:
There are all sorts of the problems with the argument that the inflation of 1933-34 was caused by expectations of fiscal stimulus. First of all, it’s completely at variance with Keynesian theory, which Kimel seems to be trying to defend. Keynesian theory says demand stimulus doesn’t raise prices when there is “slack,” and there has never been more slack in all of American history than in 1933.
The problem for Sumner is that Keynesian theory is merely an extension of good old fashioned Adam Smith. Prices depend on supply and demand. You can have a good or service go up in price locally even as it goes down everywhere else.
As I noted in my earlier post, and he quoted:
The immediate effect was to convince factories they’d be running down their inventories… After a bit of time, the public realized FDR wasn’t planning just a one-off, but rather a sustained program of purchases of industrial items. That led them to start using some of their idle capacity, which meant not just selling the fixed amount that was in inventory. The rate of price increases thus dropped.
Which of course, is very consistent with the timing of events.
None of this is to pick on Sumner. There’s a whole cottage industry dedicated to advancing a story that government spending cannot have a positive effect on the economy during recessions or depressions. The problem for those trying to advance that story is that government spending does seem to correlate with positive effects during those periods. So alternate theories are proposed, and have been proposed for decades. And those theories often make a lot of sense… until you take a close look at the data.
Not taking a side, mind you, but Sumner says in the criticizing piece in a comment below yours that FDR started devaluing the dollar in April of ’33, not October. He doesn’t cite anything for this, just that he could “recite the macro data in [his] sleep”.
“And those changes (in gold prices) were caused by explicit statements and actions by FDR. Not by fiscal stimulus, which would be expected to appreciate the dollar”…because I, Scott Sumner, say so. Sorry, but causality is a tricky enough subject, without allowing “’cause I say so”.
Not that the writers of Wikipedia can recite macro data in their sleep or anything, but the Wikipedia history of the US gold standard includes the following:
“Commercial banks also converted Federal Reserve Notes to gold in 1931, reducing the Federal Reserve’s gold reserves, and forcing a corresponding reduction in the amount of Federal Reserve Notes in circulation. This speculative attack on the dollar created a panic in the U.S. banking system. Fearing imminent devaluation of the dollar, many foreign and domestic depositors withdrew funds from U.S. banks to convert them into gold or other assets.
“The forced contraction of the money supply caused by people removing funds from the banking system during the bank panics resulted in deflation; and even as nominal interest rates dropped, inflation-adjusted real interest rates remained high, rewarding those that held onto money instead of spending it, causing a further slowdown in the economy. Recovery in the United States was slower than in Britain, in part due to Congressional reluctance to abandon the gold standard and float the U.S. currency as Britain had done.
“Congress passed the Gold Reserve Act on 30 January 1934…”
So the data Sumner is reciting in his sleep may be something he dreamed. So there were factors at work other than “defacto” anything on the part of FDR.
Also worth noting is that Sumner wants this to be a story about gold. There is certainly some overlap between those who see gold as the ultimate measure of value and those who see FDR’s policies as failed, no matter what the data may show. I don’t know that Sumner is a gold-bug, but we don’t need to accept that the value of gold is the central issue in price-stability during the 1930s simply because Sumner makes in the center of his analysis. Sumner say “gold” but the price data say otherwise. This discussion needs to remain what it was at the beginning, a contest between a gold story and a Keynesian story. Sumner seems to want to make it a “gold/not gold” debate, while Keynes is left out.
FYI, I posted this comment at Sumner’s place just a moment ago:
You know, before this I had never heard of any devaluations in 1933. I’ll admit, this isn’t my area of specialty, but I’m generally pretty good at finding data. I wonder how much real effect these devaluations had on any organization’s business. I note that the Fed’s own Banking and Monetary Statistics which show the value of monthly gold bullion set the price of gold for each month up to January 1934 at 20.67 an ounce. (See bottom paragraph, right column, page 522 here:http://fraser.stlouisfed.org/publications/bms/issue/61/download/134/section14.pdfI’ve seen the same thing in a number of other Fed and Treasury docs over the years, which explains why I never heard of this until now.) And if there are any organization that has an incentive to get that the value of gold right, its the Fed and the Treasury.
I’m having a great deal of trouble finding any series that shows something other than what the Fed used as the price of gold. (I.e., if you have access to such a series, please provide a link.)
That said, I’m assuming that, as you (and Jesse Jones) pointed out, FDR did devalue the currency. Let’s further assume that it had an actual effect, as opposed to being an announcement that was completely ignored. Now, assuming that, you actually have the opposite problem…. the effect doesn’t make sense.
Consider… I reread the Jesse Jones piece more carefully. He indicates that when he and Morgenthau had been called, the price of gold was at 29.01.
Now, 20.67 to 29.01 is a change of 40%. The price would go on to increase another 20% (to 35) by Feb 1934.
And yet by this time (i.e., by the time the price was at 29.01) the PPI had increased by 17%. It would increase by another 3% by Feb 1934. In other words, whereas the first set of devaluations were twice as large as the remaining set, the equivalent first set of PPI increases were six times as large as the last set of PPI increases.
If these devaluations had a real effect and were the cause of the changes in the PPI, why is it that the Jesse Jones / Morgenthau group of devaluations had such a small effect on the PPI relative to the earlier devaluations?
“Devaluation” here is used to mean a higher price of gold, rather than a lower dollar relative to other currencies. That is a very gold-centric view of things. I understand that, for a period which the gold standard was in use and in flux, that is a fairly natural use of the word. It is not, however, good to blithely adopt all the assumptions that go with that particular use of the word. The UK had already left the gold standard, and was a major US trading partner. What impact did the policies and economic conditions of US trading partners have on US prices, separate from the US policy toward the gold standard? Again, insistence on looking at the issue from the perspective of changes in the dollar price of gold ends up making this a story about gold, even if that’s the wrong story. The way to go about this that doesn’t skew the analysis toward “gold/not gold” is to look at other factors, as well as gold prices, and see which fits best. Domestic demand. Wages. Trade prices.
Oh, and trade patterns. Standard textbook analysis has increased trade leading to lower inflation. There was a decrease in trade among developed nations in the 1930s. If we saw a surge in wholesale prices, does the timing of reduced trade volumes fit well with the rise in wholesale prices? If so, the gold-bug story is weakened. If, instead, trade had and prices reacted in something other than textbook fashion, that would help rule out a trade explanation for US domestic price swings.
Now, about those official gold prices. The Fed and Treasury would have an institutional reason for reflecting official prices. They were, in fact, charged with maintaining the official price. What the run-on-gold story suggests is that the volume of trade in gold, interest rates and exchange rates should be considered to see whether the official price is a good guide to what was actually going on. Check gold and FX prices outside the US to see if there was likely to be a big arbitrage, suggestive of a shadown price much different from the official price.
More from Wikipedia:
“In early 1933, in order to fight severe deflation Congress and President Roosevelt implemented a series of Acts of Congress and Executive Orders which suspended the gold standard except for foreign exchange, revoked gold as universal legal tender for debts, and banned private ownership of significant amounts of gold coin. These acts included Executive Order 6073, the Emergency Banking Act, Executive Order 6102, Executive Order 6111, the Agricultural Adjustment Act, 1933 Banking Act, House Joint Resolution 192, and later the Gold Reserve Act. These actions were upheld by the US Supreme Court in the “Gold Clause Cases” in 1935.”
The “Gold Clause Cases” ruling came after the Gold Reserve Act, so may not have had much effect on policy. Don’t know.
I’m amazed at Sumner’s commenters accusing you of being partisan. You are one of the most empirically driven bloggers on the web and rarely concern yourself with politics. Seems the right are simply so sued to bias that neutral sources are obviously liberal propaganda!
“And yet… the graph shows very clearly that prices started to rise when FDR came to Washington with spending plans”
Not really. The graph in your post is of the year-on-year percentage change in the price level, not the price level itself. The price level didn’t really begin rising (except for a one-off increase in the PPI between February and March) until after FDR took office, as Sumner argued. I’ve put both the PPI and CPI in a graph and set March 1st, 1933, as the base period (ie. a value of 100) to clearly show this:
http://research.stlouisfed.org/fredgraph.png?g=3mY
To download the data, go here and click “Download Data in Graph”: http://research.stlouisfed.org/fred2/graph/?g=3mY
You can see the PPI reached its lowest point around February. There was then an increase over the course of February, but after that the PPI stabilized. It didn’t begin rising continously until sometime between April and May. The CPI reached its lowest point around March and stabilized. It didn’t begin rising continuously until sometime between May and June. This seems more supportive of Sumner’s theory than yours.
Come on Mike,
FDR annouced the plan to depreciate the dollar against gold in April of 1933 (against the advice of the majority of advisers and against the consensus view on Wll St.).
The economy was still in deflation at the time, industrial production was still falling and stock prices, remained very close to thier all-time lows (which were seen in the middle of 1932). On the day FDR announced that the dollar would be depreciated, the DJIA posted it largest ever single-day percenatge increase. Industrial production had fallen steadily from October of 1932 to March of 1933. Following FDRs surprise annoucement, indsutral production postsed its largest ever 4-month increase, surging 55.3%. CPI troughed in March of 1933. In the four-months following the annoucement, consumer prices jumped along with IP at a blistering annualized pace of 12.4% (4% not annualized). Stock prices also showed thier larest ever four month increase, rising 80.3%.
Wow. Yes, let’s attack Sumners’ commenters’ attacks on Mike. That should distract us from the fact that Mike just picked the topic of *monetary policy in the Great Depression* in a fight against an academic expert on *monetary policy in the Great Depression*, and used the phrase “very wrong”. “empirically driven” indeed.
kharris,
I had a brief exchange of e-mail with Sumner. He shared with his data source: the Annualist Index of Commodity Prices. He apparently collected those manually from old hard copies so there may not be an on-line analog. In my last e-mail message to him I’ve asked him for permission to put the table on-line, or if he might do so at his blog and we could then link to it.
I’ve also asked him what the transmission mechanism was for devaluation in, say, June of 1933.
In brief, I don’t think he’s making up data. I do question whether the data means anything, however. Prices aren’t relevant if nobody is using those prices.
Staidness,
I did use the phrase. And Sumner accused Skidelsky of cherry-picking (see my previous post). I am currently standing by my phrase.
As noted upthread, he and I had a brief exchange of correspondence this afternoon. I have asked him if we could post the index he uses as evidence that the dollar was devaluing. My problem with the index is this: he views it as a price that is relevant. I view it as a price that isn’t relevant. I’ve noted (link upthread) that the Fed valued gold assuming prices were unchanged through January 31, 1934. I’ve seen other government sources do the same.
What we come down to is this. Sumner may be an expert, but he is insisting that changes in a specific index show the dollar devaluing relative to gold. But the Fed and others who had a strong interest in getting the value of gold right were using a price that was fixed through all of 1933.
I have in the past stated things that go against conventional beliefs, so I am sympathetic to Sumner in principle if he’s going to do the same. But there is a difference… this isn’t a theory issue, its a practice issue. If he feels that the “true” price of the dollar relative to gold started to change in April, its important to find examples where players at the time did take changes in the price into account. If the Fed and everyone else was using the fixed price, then that by definition was the “true” price.
inyourhouse,
I could have phrased things a bit more precisely, but I did put up the graphs I did for a reason.
If a plane has stalled, the point where things have improved is not when it is starting to rise again. Its when the engines have finally gone back on during the dive.
Prices were in free-fall. They started stabilizing a bit before FDR took office. Then they began to rise.
Gregor Bush,
FDR announced many things in the first 100 days. Nobody knew what was going to stick and what wasn’t.
Which brings us back to what I’m trying to state: the program to increase spending was there. The devaluation of the dollar relative to the dollar is there according to an index Sumner is using. But contra that index, there seems to be evidence to suggest that in practical terms, the dollar did not devalue against gold until 1934. If so, just about every point you bring up is due to something else. I’ve suggested a something else that fits the data, and whose existence is supported more widely.
Staidness, inyourhouse, Gregor Bush,
I think I know how to say what I’m trying to say. Sumner has a series. I am asking whether that series is the equivalent of the MSRP on a car, a price that nobody pays.
With temporal distance, its sometimes not easy to tell whether something is real or not. In 100 years someone might reach a number of conclusions about the housing market from reading press releases or peraps even data provided by the National Association of Realtors in recent years. Those conclusions will be well supported, but they will be wrong.
Your comment is awaiting moderation.
November 15, 2011 at 9:06 pm
Hello Mike! Great blog I have now bookmarked it. I have been reading Sumner’s Money Illusion which I find interesting though I do differ with him on the efficacy of fiscal policy.
For my attempts to gloss Sumners and the monetary debates please see
http://diaryofarepublicanhater.blogspot.com/2011/11/ngdp-need-to-focus.html
To boil it down, though, what exactly is your difference with Sumner here about? He credits the turn around in prices mostly to FDR getting us off the gold standard whereas he claims that the National Economic Recovery Act at best did little good at worst even retarded the strong recovery precipitated with the currency devaluation.
For clariy let me quote is answer to this post “He insists that FDR’s dollar depreciation program began in October 1933, even though all economic historians agree in began in mid-April 1933, when the exchange rate for the dollar began declining (against gold and against other currencies.) He insists prices began rising before FDR took off, which is not true. He presents a graph that he claims shows prices rising before FDR took office, but his graph shows inflation rates, not the price level. In fact, the graph actually supports my argument that inflation didn’t turn positive until after FDR took office. There’s a difference between the rate of inflation and the price level.”
http://www.themoneyillusion.com/
This is an age old disagreement-was it monetary policy that led to the strong recovery or fiscal. I consider myself a Keynesian my instinct probably says a bit of both. But Sumner like most monetarists place it mostly with monetary policy. What is your position to quantify it, are you saying what FDR did with gold wasn’t a big part of it or simply that his fiscal stimulus was also important?
Out of 100 percent what percentages respectively would you ascribe to fiscal and monetary policy… LOL
It seems Scott Sumner is digging even deeper. He wrote another post (http://www.themoneyillusion.com/?p=11919)
Here’s the comment I left:
1. “He insists prices began rising before FDR took office off, which is not true.”
PPI (all commodities) was 10.3 February 1. That happens to be the lowest point in the series, a series that begins in 1913. It was 10.4 in March 1, 1933. FDR took office March 4, 1933. (http://research.stlouisfed.org/fred2/graph/?id=PPIACO)
Explain how my statement is not true. I’m genuinely curious. Yes, it was only inching up by then, but then, after the dramatic recent drop that was in itself a pretty big deal. And it happened before the devaluation.
2. ” There’s a difference between the rate of inflation and the price level.” And yet the price level had bottomed out before FDR took office regardless of how you look at things.
3. As to what economic historians have said… well, I am definitely not an economic historian. But I can use google.
I have noted (I’m not going to put up links again – they’re in my earlier post if you need them) that in documents I have seen from the 1940s, the Fed was valuing monthly gold reserves for 1933 using the price of 20.67. Those same documents seem to be written by people who were under the impression that the price remained 20.67 until January 31, 1934.
Sumner and I have had a brief correspondence this afternoon. He kindly shared a piece of a manuscript he is writing that contains a table that has in it the “Annualist Index of Commodity Prices.” The index shows falls from 815 in mid-April (1933) to about 650 the following February, which indicates “falling prices in terms of gold.”
He apparently collected those manually from old hard copies and is unaware of any electronic versions. In my last e-mail message to him I’ve asked him for permission to put the table on-line. Or he can put it up here – its possible he already has somewhere, but a quick google search hasn’t turned it up.
But in any case, we come down to some Fed documents claiming prices were fixed until January 31 1934 v. an index. Assuming the index is correct, why would the Fed et al be under that misimpression?
A few random thoughts by someone who is not an economic historian but wondering about that question….
a. I grew up in South America at a time when the official price of a given currency (in dollars) was usually a work of fiction that had almost no effect on anyone’s behavior but did set a baseline for certain government transactions. Changing the official dollar price of gold can allow more dollars to be printed even if nobody else acts on the official price change.
b. there is such a thing as a manufacturer’s suggested retail price (MSRP) in autos and other goods. There apparently are good reasons for some products to have MSRPs even if nobody pays that price.
c. The existence of a series doesn’t mean the series is correct. A hundred years from now someone might choose to analyze the real estate market over the past ten years or so analyzing “days on the market” data from the MLS. No doubt that someone will have their conclusions well-supported by the data set they are using. But the results will still be […]
Mike Sax,
I’ve had many posts indicating I believe monetary policy matters. Its also in the book Presimetrics. I would be shocked if it didn’t matter in the 1930s.
That said, I have a hard time seeing how the government stepping in and spending money at a time when nobody else was willing to do it didn’t have a big effect.
As to the percentage… I’ve been playing with some numbers looking at the correlation between the growth rate of a function of nondefense spending and the growth in real GDP. That correlation is extremely high through 1938. I hesitate to write it up… I want to get back to the “kimel curve” stuff I was working on which I find more interesting, but I keep thinking if I write the right thing it will go away…
You’re right, I was trying to ‘distract’ you rather than simply offering an observation. Did I hit a nerve?
Just for grins and giggles… I found another source this morning. The 1936 Statistical Abstract of the United States (http://www2.census.gov/prod2/statcomp/documents/1936-05.pdf), table 230.
The data clearly doesn’t originate with the Census, but rather with the Fed and Treasury. Footnote 1 is completely unambiguous about what those agencies believed the price was in June of 1933 (a few months after the devaluation started according to Sumner), and the footnote goes on to state precisely when the gold price changed and by how much. And it doesn’t fit Sumner’s story line at all. I have yet to find a single relatively contemporaneous source that treats the price of gold as actually changing. Yes, FDR made pronouncements, but I’m starting to see it the same way as GW changing the status of waterboarding from torture to something else – a fiction that greased some government activities that were going to occur anyway (and would have gone forward with a different excuse had that one failed) rather than as anything substantive.
Hey Mike! Being new to your blog haven’t read your other stuff yet. Is Presimetrics your work? What is the Kimel curve?
Mike,
Yes FDR annouced alot og things in his first 100 days, but the record-setting surge in stock and commodity markets came on the day he surprised everyone by annoucing that the US was going to de-value the dollar against gold. That’s powerful evidence that the annoucement had an immediate impact of expecations of the future price level.
“I’ve suggested a something else that fits the data, and whose existence is supported more widely.”
I don’t think that’s right. Most economists who study the Depression such as Ben Bernanke, Barry Eichengreen and Christina Romer point to the central role of monetary expansion via leaving the gold standard as the most important policy step that led to recovery.
As Brad DeLong pointed out, the six largest economies in the world began to recover from the Depression in the exact order that they devalued against gold. The odds of achieving that rank order purely by chance are 1 in 120.
Sure, if you take a US history class, the professor will tell you that the economy recovered becasue of FDRs spening plans. The dollar devaluation probably won’t even get mentioned. But that only becasue history professors have at least a vague understanding of fiscal policy and almost zero understanding of monetary policy.
Did fiscal policy help? Yes.
If the US had targeted $50 instead of $35 (like George Warren wanted) would the price level have risen more quickly? Yes.
If the expected future price level had risen more quickly would the recovery have been even stronger? Yes.
If the US had left gold in 1930, would there have been a Great Depression? No.
Here’s Irving Fisher writing in the fall of 1933:
“Those who imagine that Roosevelt’s avowed reflation is not the cause of the recovery but that we had “reached the bottom anyway” are very much mistaken…If reflation can now so easily reverse the deadly downswing of deflation after nearly four years, when it was gathering increased momentum, it would have been easier still, and at any time, to have stopped it earlier. It would have been still easier to have prevented the depression almost altogether.”
OK Mike I just looked up the presimetrics blog, etc. So you did write a book. Very cool. As regards what I was already able to see about the Kimel curve, I can see why you’re more interested in that than arguing about 1933 with Sumner. The idea of finding a relationship between the top marginal rate and maximizing GDP growth is more interesting. I will check that out.
Mike Sax,
I was co-author of the book. The website Presimetrics is just a nice place to store a lot of the posts I’ve been putting up here at Angry Bear for about the past two years. (For earlier posts I wrote at Angry Bear you have to work through the archive at Angry Bear.)
The Kimel curve is pretty simple. It also infuriates a lot of people. Ironically, folks who will insist that the Laffer curve applies to the US economy and cannot in a zillion years manage to fit it there will insist something is wrong with the Kimel curve.
As always in those posts, the data is publicly available but I am also happy to provide my spreadsheets if desired.
Gregor,
Simply put, I haven’t had an opportunity to look at all the data that Sumner has examined over the years, but broadly, I have no disagreement over when different things started picking up in 1933. I am only pointing out that the devalutions that he sees weren’t really there. I am sending Dan a post in a few minutes that will expand on that point. I imagine it will go up later today or tomorrow.
There is this word that gets over-used in economics, but this may be the place to pull it out – “robust”. Often, it means “I like the math I used to reach this conclusion” but sometimes, in better writing, it means “jigger the dates, use any credible data source, impose or remove limits all you want, this result is solid no matter what.” Sumner is using a data series to which others don’t have access, which contradicts other data series about gold prices, and insists that his series is the right one, that conclusions that can be drawn from his series but not from others are the true conclusions. That ain’t what “robust” means when used rightly.
One of your commenters argues that Sumner is an “authority” and suggests you shouldn’t question him. That kinda comment needs a warning label marked “arguing from authority”. As I understand it, Sumner is not a vastly skilled technician who follows the data where ever it leads. Isn’t Sumner a long-standing critic of government intervention, Keynesian and FDR critic? Even if we allow arguing from authority in general, we need to have a separate category for “arguing from my priors”.
And, in one last, probably futile attempt, I urge you to derail Sumner’s efforts to make this all about gold. (“Lookee at this obscure series on gold prices I put together…”) Sumner’s initial claim that Keynes/FDR explanations fail is not substantiated, but he has walked away from that and gotten into quibbling about the minutia of gold prices. That’s like allowing Hensarling to make the deficit discussion all about job-killing tax hikes. Pretty soon, there’s no place left for a discussion of reality. That’s why they do what they do.
Mike,
As you rightly said, there were a ton of things that FDR did. So, to pick out a particular thing and point out that as the driving factor is a stretch. Be that as it may, the confusion comes from the use of the word “devaluation.” On April 19, FDR took the US off the Gold Standard. Yes, the market rallied powerfully, but to put things in perspective, it was not one of the 10 biggest single day % point rallies in the Dow, let alone the biggest.
However, prior to that FDR took a number of actions that also constrained the gold standard, including the March 5 order that announced a bank holiday and embargoed exports of gold and silver and the March 9 emergency banking act (I think) made private gold holding illegal. All of these were superseded by the act of June 5.
Note that when the 4-day bank holiday announced by FDR as soon as he took office ended and banks reopened on March 12, deposits poured into banks. BTW, March 15th was the biggest single day rise in the Dow. It is not surupising that, once the run in banks was stopped, normal credit required for regular business activity could resume and prices rose simply because businesses no longer had to sell things on fire sale simply to raise cash. In fact, before FDR took office, barter among ordinary people had risen due to the extreme shortage of cash. So, yeah, bank runs are not good, credit is needed for normal activity. In that sense, moentary policy worked. But if anyone thinkgs we are in a similar situation or that it holds any parallels today, I think they are fooling themselves.
The Fisher quote uses the term “reflation”. That term does not differentiate between causes of reflation, and so doesn’t help to distinguish between fiscal policy-driven and monetar policy-driven results.
Declaring that history professors’ views should be ignored because they don’t know what some other expert knows builds in the result you are arguing for, without demonstrating that result to be true. (This is “begging the question” in its original meaning.)
Mike’s work comes down, most generally, to looking at data. Arguments from authority (even if the argument from authority misses the mark, as is the case with the Fisher quote) is not an answer to the implications of the data.
kharris,
OK, I won’t “argue from authority” anymore.
The two biggest days for the stock market in 1933 were March 15th (+15%, the day FDR announced the government would guarantee bank deposits) and April 19th (+9.4%, the day FDR announced that the US would be going off gold).
Both policies had the effect of dramatically loosening monetary conditions. The first by stopping the run on the banking system and therefore stopping the reverse multiplier effect that was occurring, whereby banks were losing deposits and being force to call loans. Going off gold loosened monetary conditions dramatically by raising the expected future piece level and dramatically lowering ex ante real interest rates and ex ante real wages. And because agents in the economy knew there was massive excess capacity, it also raised expectations of future real output.
The trough of the CPI was in April of 1933 and the trough of IP was March of 1933. From April to July industrial production, stocks prices, commodity prices and consumer prices surged.
Japan went off gold in 1930 and started to recover despite the global Depression gathering steam. Britain and Sweden went of gold in 1931 and started to recover despite the rest of the world continuing to plunge deeper into the Depression. France and Belgium stayed on gold until 1936 and remained in the depression despite the fact that the US (and the rest of the world) was recovering. When they left gold they started to recover.
Here’s Brad Delong’s chart on the world’s six largest economies and when they left the gold standard:
http://delong.typepad.com/sdj/2009/03/the-earlier-you-abandon-the-gold-standard-and-start-your-new-deal-the-better.html
Are you finding any of this convincing? Oh, and by the way, Fisher was clearly talking about monetary expansion.
The chart is of a weekly PPI, but this is the first I have ever hear of anyone constructing a weekly WPI.
Daily and/or weekly commodity price indices exist and I wonder if he is using some commodity price index rather than the PPI.