Evan Soltas on the 1920s housing “bubble.”

Here’s an interesting new Evan Soltas piece in Bloomberg:

The U.S. economy slips into recession. The stock market takes a tumble as heady expectations for future growth cool. Interest rates fall as the Federal Reserve quickly trims the discount rate in hope of cushioning the business cycle.

The low interest rate environment sets off a massive wave of home construction and an asset bubble in real estate. By the time the Federal Reserve takes action, the boom is completely out of control. Bank balance sheets and household savings have become dependent on the profound mispricing of real estate and other equity holdings.

The heedless extent of leverage makes the financial system extremely vulnerable to capital losses. As the housing bubble implodes, it pushes the economy into a long, deep recession.

This is the economic story of the last decade — and of the 1920s.

After a sharp deflationary recession at the end of World War I, the newly created Federal Reserve slashed interest rates, setting off a housing bubble of such an incredible scale that it dwarfs its recent counterpart. When the bubble ended, what seemed to be a calm and contained contraction turned violent, culminating in the macroeconomic implosion of the Great Depression.

It sounds a lot like the recent crisis, but I also see some important differences:

1.  Monetary policy was not expansionary during the 1920s.  And indeed Evan doesn’t present any evidence suggesting it was easy.  Take whatever criteria you like, starting with interest rates.  Evan mentions that interest rates fluctuated between 3% and 7%, but fails to mention that the 1920s began with deflation (1920-21), and then saw level prices for the rest of the decade.  With inflation expectations near zero (typical of a gold standard regime) real rates ranged for normal to high. The monetary base was flat during the 1920s, a decade that saw rapid growth in population and RGDP.  Even if you take sub-periods where the base increased, it was certainly not expansionary compared to most other periods of history.  NGDP rose slightly during the 1920s, but fell sharply in per capita terms (which is very unusual.)  And there was no “bubble” in real estate prices, Soltas mentions a price bubble in Manhattan, which did occur, but there was no significant nationwide change in either real or nominal house prices during the 1920s.

2.  Although money wasn’t easy, it’s still possible that the huge rise and fall in housing construction somehow caused a depression to begin in 1929.  But it’s hard to see how that would have occurred.  Suppose NGDP had remained stable after 1929, instead of falling by 50%, does anyone seriously believe a housing decline would have caused a depression?  I suppose one could use a re-allocation story, but then other sectors should have been booming, and they were also declining. Furthermore, a substantial amount of decline took place between the peak of 1926 and mid-1929, and yet the economy continued booming. Previously I’ve pointed to the fact that the US economy didn’t fall off a cliff between January 2006 and April 2008, despite a huge decline in housing construction.  But the 1926-29 period is even more problematic for bubble worriers, as while 2006-08 saw an economic slowdown, 1926-29 was actually a boom period.  It’s not obvious why the boom could not have continued into 1930.

3.  So the problem wasn’t tight money, and the problem wasn’t the decline in housing construction.  Perhaps the decline in housing construction somehow caused a banking crisis, which dragged down NGDP (due to the gold standard.)  But why would a decline in housing construction have caused banking crisis? Even worse, in 2007 we saw the banking system come under great stress, despite the economy not even being in recession.  Now contrast that with the period from August 1929 to October 1930, which saw a recession even deeper than 2009, and no banking crisis at all!  Why the difference?  I can only speculate, but perhaps there were two key differences.  In the 2000s banks made the sort of highly risky mortgage loans that banks did not make in the 1920s (although loan quality fell off a bit late in the 1920s.)  And second, in the 2000s there was a huge increase and then decrease in home prices.  Because there was no price bubble in the 1920s, there was no banking crisis in the first 14 months of the Depression.  Instead, tight money by the Fed, the BOE, and the Bank of France created world-wide deflation and drove NGDP much lower during 1929-30.  The banking crises only began to develop when falling NGDP drove nominal incomes much lower.

4. Now it is true that the fall in NGDP did eventually get so bad that a banking crisis developed, but you’d expect that with any severe decline in nominal incomes.   A very mild banking crisis occurred in Tennessee in November 1930, and then a more severe one developed all over the world in mid-1931.  That’s what you’d expect in a world on non-indexed debt and rapidly falling NGDP.  And that led to gold and currency hoarding, which drove NGDP still lower.

Of course in pointing to the fall in NGDP I’m not telling Evan anything he doesn’t already know, and indeed has written eloquently about.  If you don’t want a collapse in RGDP, tell the central bank not to allow a collapse in NGDP.

And make sure you don’t screw up the supply-side of your economy with disincentives to produce.


Tags:

 
 
 

25 Responses to “Evan Soltas on the 1920s housing “bubble.””

  1. Gravatar of Evan Soltas Evan Soltas
    30. September 2012 at 11:23

    Thanks for your notes. I’ll come back and work on a response, but first I want to note you have my last name spelled wrong.

  2. Gravatar of Lorenzo from Oz Lorenzo from Oz
    30. September 2012 at 12:58

    Point 3: “The problem wasn’t tight money”. Is that what you meant to write?

  3. Gravatar of Bonnie Bonnie
    30. September 2012 at 13:36

    So what about the S&L crisis and the stock market crash of 1987? We had a recession in the early 90’s… maybe it was long and variable lags at work?

  4. Gravatar of Ryan Ryan
    30. September 2012 at 13:50

    It sounds a lot like the recent crisis, but I also see some important differences:

    1. Monetary policy was not expansionary during the 1920s.

    Again and again you tell us that monetary policy was tight leading up to the current crisis, as evidenced by the fact that interest rates were low. Now you’re telling us that money was loose in the recent crisis. (At least, that’s what I infer by your contrasting the 1920s “not expansionary” policy as an “important difference” it has with “the recent crisis.”)

    Maybe I’m just not following you.

  5. Gravatar of ssumner ssumner
    30. September 2012 at 14:43

    Sorry Evan.

    Thanks Lorenzo, I changed it.

    Bonnie, No, the 1987 crash was followed by a big economic boom.

    Ryan, Good point. I meant to say that money was easy by conventional standards in the 2000s, but not by conventional standards in the 1920s. My mistake.

  6. Gravatar of Jonathan M.F. Catalán Jonathan M.F. Catalán
    30. September 2012 at 15:10

    The post-WWI deflationary recession Evan mentions is the deflation during the early 1920s (the recession of 1921-22).

  7. Gravatar of Morgan Warstler Morgan Warstler
    30. September 2012 at 16:21

    “If you don’t want a collapse in RGDP, tell the central bank not to allow a collapse in NGDP.

    And make sure you don’t screw up the supply-side of your economy with disincentives to produce.”

    Spoken like a true egghead who doesn’t know his place.

    Let me fix it for you:

    Bend to the will of businessmen. Accept their dominant social status; understand that most are not able to do the truly hard stuff that creates jobs for everyone else. Those that aren’t able to do the hard stuff will be slightly lower status, sometimes they will become economists.

    And oh by-the-way, the Central Bank needs to keep a LOW stable price level that generally rewards savers more than they have been rewarded in the past.

    Getting econbloggers to remember who they work for shouldn’t be this hard.

  8. Gravatar of Dirk Dirk
    30. September 2012 at 16:49

    Have you ever read Thomas Wolfe’s You Can’t Go Home Again ? I wonder how much his narrative of the housing bubble & crash in his home town of Asheville, North Carolina (which, according to Wiki, was one of the worst hit cities in the depression) influenced the popular narrative of the “housing bubble” of the 20’s. As he tells the story, his home town was a microcosm of the housing boom-bust which had occurred nationally. But, in fact, as you mention, it wasn’t. His home town was a rather exceptional case — or at least an extreme one. But his novel was very popular in the 30’s and I wonder if his narrative didn’t seep into the national consciousness to the point that it eventually became (popularly) The Narrative. In other words, perhaps if Thomas Wolfe had been from a different city, a different legend would have been passed on to us regarding the cause of The Great Depression.

  9. Gravatar of Bonnie Bonnie
    30. September 2012 at 16:54

    I was being snarky. I understand that the difference between 1929 and 1987 is competent monetary policy. And I wish Mr. Soltas wouldn’t feed the trolls.

  10. Gravatar of JimP JimP
    30. September 2012 at 16:57

    An Economist NGDP poll:

    http://www.economist.com/economics

  11. Gravatar of Major_Freedom Major_Freedom
    30. September 2012 at 17:36

    Monetary policy was expansionary during the 1920s. Consider the facts. Real productivity was quite high during the 1920s. During the 1920s, stock prices quadrupled. Durable goods increased 160 percent. Non-durable goods increased 60 percent. Manufactured foods rose 48 percent.

    Yet prices did not significantly rise (the Fed adopted a “stable price” policy). The only way this could have happened is if aggregate nominal demand kept up with the aggregate real supply.

    How in the world can all that productivity and stable prices NOT imply anything other than an expansionary monetary policy?

    Aggregate money supply (especially credit expansion) is a decisive measure of monetary policy tightness and looseness, and according to this metric, the aggregate money supply growth during the 1920s suggests expansionary monetary policy.

    You said “Take whatever criteria you like”, but then you only mention interest rates, monetary base, and NGDP. You are ignoring aggregate money supply.

    From 1921-1929, the money supply increased by $28.0 billion, a
    62% increase. This is an average annual increase of 7.7%.

  12. Gravatar of Major_Freedom Major_Freedom
    30. September 2012 at 18:10

    1920s money at a glance.

  13. Gravatar of Peter N Peter N
    30. September 2012 at 18:49

    Major Freedom has the right of it this time.

    While he was alive, Benjamin Strong dominated the Fed. He believed in stable prices and, in effect, subverted the normal balancing mechanisms of the pre-WWI gold standard. There never really was a real post WWI gold standard, since many of the central banks manipulated it for various reasons and some like England tried to restore pre-WWI parity creating distortions.

    When Strong died, believers in the real bill’s doctrine took over and were appalled at the Fed’s financing what they saw as speculation. They tightened money by, among other things, jawboning the banks through the dominance of the NY Fed to make them stop using the discount window (threatening to audit them when they do is a rather strong disincentive), raising the price of call money and publicly disavowing any responsibility or intent to use monetary policy to stabilize prices.

    “{Adolph C.] Miller (1935: 453) claimed critically that Strong’s policies in the presence of stock market speculation ‘proved to be unequal to the situation . . . in this period of optimism gone wild and cupidity gone drunk.’ The Federal Reserve Board’s ‘anxiety,’ he continued, ‘reached a point where [the Board] felt that it must assume the responsibility for intervening . . . in the speculative situation menacing the welfare of the country.’ Consequently, on February 2, 1929, the Board sent a letter, crafted mostly by Miller, to all the Fed Banks stating that the Board had the ‘duty . . . to restrain the use of Federal Reserve credit facilities in aid of the growth of speculative credit.” To accomplish this end, the Board initiated ‘he policy of ‘direct pressure’ [that] restricted borrowings from the federal reserve banks by those member banks which were increasingly disposed to lend funds for speculative purposes’ (p. 454).

    A bank not able to pass the “direct-pressure test” could
    not borrow from a Federal Reserve Bank at any rate, no matter how much “good” paper it had or how badly it needed “credit” to meet deposit withdrawals. To such a client bank, discretion by authority substituted for the Fed discount rate to ration Federal Reserve “credit.” Ironically, the policy was completely contrary to the positive prescriptions for Fed Bank discounting set forth in the Federal Reserve Act.


    At the same time that Fed policymakers refused to provide relief to member banks, gold in Fed Banks was piling up. By August 1931, Fed gold had reached $3.5 billion (from $3.1 billion in 1929), an amount that was 81 percent of outstanding Fed monetary obligations and more than double the reserves required by the Federal Reserve Act.”

    http://www.cato.org/pubs/journal/cj28n2/cj28n2-13.pdf

    Unsurprisingly as 1929 progressed, the economy began to slow down, and the crash was the natural result. Sort of like Wile E. Coyote suddenly discovering he is walking on air.

    The Fed’s policies combined with congress’s obsession with paying down the national debt aborted a series of attempted market rallies, until the financial economy dragged down the production economy as money remained tight.

    The subsequent failure of Credit Anstalt and the subsequent major German bank failures combined with British-French conflict over the gold standard then finished the job of guaranteeing a depression. It was just a question of how bad idiotic policies could make it. The answer, was very bad. Catastrophic if you include the minor matter of causing WWII.

    As for housing, the need to roll over short term mortgages was definitely an aggravating factor in the face of a credit contraction. You don’t need much of a real estate boom to have a real estate bust, if you have the Fed to help with the busting.

  14. Gravatar of Major_Freedom Major_Freedom
    30. September 2012 at 19:26

    This blog habitually insists that interest rates and the monetary base are not valid factors for judging the tightness or looseness of monetary policy, and yet, when defending the “monetary policy was not expansionary during the 1920s” judgment, we see interest rates and the monetary base being cited as factors, which of course implies they are valid ones.

    NGDP can be used to judge monetary policy, as long as it is understood that “stable” or “gradually rising” NGDP can be associated with expansionary monetary policy, if we consider the counter-factual, undisturbed laissez-faire NGDP to be anything lower than what actually occurred under central banking.

    It is wrong to presume that a free market in money would have a constantly growing NGDP. A period of stable, or even declining NGDP may be associated with highly expansionary monetary policy, if the same economy with no central bank would have resulted in outright declining NGDP.

    Just like it would be wrong to presume that an individual firm, person, or collection of competing companies in a given sub-country sized geographical territory should earn a constantly growing income, so too would it be wrong to presume that a country’s population of income earners should earn a constantly growing income.

    NGDP targeting cannot be the proper standard for deciding whether or not a given NGDP is tight or loose. It’s circular logic.

  15. Gravatar of TheMoneyIllusion » Evan Soltas on the 1920s housing “bubble.” | Online Study Zone TheMoneyIllusion » Evan Soltas on the 1920s housing “bubble.” | Online Study Zone
    30. September 2012 at 22:45

    […] the rest here: TheMoneyIllusion » Evan Soltas on the 1920s housing “bubble.” var addthis_config = […]

  16. Gravatar of An Interesting Measure of Prices An Interesting Measure of Prices
    1. October 2012 at 06:42

    […] and Scott Sumner, according to this metric at least, there was a big surge in (price) inflation during the 1920s, […]

  17. Gravatar of John Brennan John Brennan
    1. October 2012 at 08:03

    With respect to Evan’s argument, I believe that the data underscore most of Dr. Sumner’s points. Below are Case Shiller Price Index data for the years in question. I include the nominal index, because that relates to the nature of the impact Dr. Sumner discusses in point 3 of his response. I also include the year over year (%) nominal and real price change (from previous year). In nominal prices, 1919 and 1920 saw significant price changes, but that was largely a response to the end of WWI. Otherwise, the highest nominal change, YoY, was 1925 at 5.21. The highest YoY real price change (accounting for deflation and inflation during this time period) shows 1921-1922 with the only double digit increase in housing price. No bubble trend can be detected with either measure. The 1920’s represented the tail end of America’s major urbanization phase that coincided with the mass adoption of electricity and automobility. Housing followed these trends. All three of these sectors (electricity, automobility, housing) were very productive during this phase and not conducive to “overinvestment” in the sense that the recent housing boom was. The banking crisis of the 1929-32 years made the housing sector worse (as the data indicate), not the other way around.

    Year CS_Nom YOY_N YOY_R
    1915 4.27 -8.19 -9.10
    1916 4.68 9.48 6.32
    1917 4.77 2.03 -9.29
    1918 5.08 6.36 -11.10
    1919 5.59 9.97 -6.68
    1920 6.12 9.60 -6.29
    1921 5.98 -2.23 -0.69
    1922 6.07 1.39 13.99
    1923 6.16 1.47 2.07
    1924 6.17 0.19 -2.70
    1925 6.49 5.21 5.21
    1926 6.23 -4.04 -7.25
    1927 6.00 -3.73 -1.53
    1928 6.09 1.49 2.66
    1929 5.96 -2.05 -0.91
    1930 5.70 -4.3 -4.3
    1931 5.24 -8.15 -1.21
    1932 4.69 -10.46 -0.44
    1933 4.51 -3.81 6.62
    1934 4.64 2.90 0.56
    1935 5.10 9.76 6.53
    1936 5.26 3.21 1.72
    1937 5.40 2.56 0.38

  18. Gravatar of John Brennan John Brennan
    1. October 2012 at 08:12

    Maybe this will look better:
    Year CS_Nom YOY_N YOY_R
    1915 4.27 -8.19 -9.10
    1916 4.68 9.48 6.32
    1917 4.77 2.03 -9.29
    1918 5.08 6.36 -11.10
    1919 5.59 9.97 -6.68
    1920 6.12 9.60 -6.29
    1921 5.98 -2.23 -0.69
    1922 6.07 1.39 13.99
    1923 6.16 1.47 2.07
    1924 6.17 0.19 -2.70
    1925 6.49 5.21 5.21
    1926 6.23 -4.04 -7.25
    1927 6.00 -3.73 -1.53
    1928 6.09 1.49 2.66
    1929 5.96 -2.05 -0.91
    1930 5.70 -4.3 -4.3
    1931 5.24 -8.15 -1.21
    1932 4.69 -10.46 -0.44
    1933 4.51 -3.81 6.62
    1934 4.64 2.90 0.56
    1935 5.10 9.76 6.53
    1936 5.26 3.21 1.72
    1937 5.40 2.56 0.38

  19. Gravatar of Doug M Doug M
    1. October 2012 at 08:56

    There was definitely a mortgage crisis in the 1930s.

    However, the mortgage market of the early part of the century was very different from the mortgage market of today.

    The predominant form of mortgage was the balloon mortgage. You paid interest monthly. At the end of 5 years, you paid down as much of the remaining principal balance as you could and hoped that the bank would extend to you a new mortgage for another 5 years. However, with banks collapsing, banks were unwilling to extend mortgage credit, even to customers who were current on their payments.

    There was no national mortgage market, only regional markets. If your local bank wouldn’t extend credit, good luck finding someone who would.

    I wouldn’t say that the housing crisis caused the great depression, or a banking crisis within. I would say that the weak banking system exacerbated a housing problem and delayed recovery from the depression.

    The government created frist the Federal Home Loan Bank system, and a few years later the Federal National Mortgage Associaton (Fannie Mae) to rectify these shortcommings — capitalize local mortage banks, standardize the 30-year level pay mortgage, and create a national mortgage market.

  20. Gravatar of Major_Freedom Major_Freedom
    1. October 2012 at 10:48

    If one defines as “expansionary monetary policy” the act of a central bank printing money to prevent “demand side” unemployment and declining output relating to a single firm, i.e. prevent “insufficient” firm level sales revenues, then to be consistent, then one would have to define as “expansionary monetary policy” the act of a central bank printing money to prevent “demand side” unemployment and declining output relating to a population of firms in an OCA, i.e. prevent “insufficient” OCA level sales revenues.

    If MMs deem themselves justified to asserting and seeking the enforcement of a minimum sales revenues for an OCA, then to be consistent they would have to deem others justified in asserting and seeking the enforcement of a minimum sales revenues for single firms, and even for individual persons.

    Oh sorry, I forgot, internally contradictory convictions are permitted in the name of pragmatism.

  21. Gravatar of Greg Ransom Greg Ransom
    1. October 2012 at 17:45

    Have you read Ed Leamer’s empirical paper on housing & the trade cycle across the 20th century?

  22. Gravatar of Greg Ransom Greg Ransom
    2. October 2012 at 07:42

    Deal with this Scott — Leamer: Housing is the recession:

    http://m.nber.org//papers/w13428

    or punt.

    I notice you like to punt — I’d recommend the punt, actually.

  23. Gravatar of ssumner ssumner
    2. October 2012 at 13:03

    Greg, A few months ago I published a paper in an economics journal that was critical of Leamer.

    http://www.degruyter.com/view/j/ev.2012.9.issue-1/1553-3832.1850/1553-3832.1850.xml?format=INT

    The fact that housing tends to decline before recessions tells us NOTHING about causation. So do stocks.

    Peter, The gold standard was manipulated both before and after WWI, so there was nothiong special about the interwar years. And everything MF says is wrong so I’d recommend ignoring his comments.

  24. Gravatar of Major_Freedom Major_Freedom
    14. November 2012 at 13:54

    Peter, The gold standard was manipulated both before and after WWI, so there was nothiong special about the interwar years. And everything MF says is wrong so I’d recommend ignoring his comments.

    You need him to ignore my comments so that you can brainwash him with yours, which incidentally happen to be often refuted by me.

  25. Gravatar of Free Banking » Booms, Bubbles, Busts, and Bogus Dichotomies Free Banking » Booms, Bubbles, Busts, and Bogus Dichotomies
    3. September 2013 at 05:47

    […] it's far from evident that he considers this something other than a merely theoretical possibility. He denies (appealing again to the bubbles-vs.-fundamentals dichotomy), that monetary policy played any part […]

Leave a Reply