Saturday, December 20, 2014

Huerta de Soto gets it Wrong on the Gold Standard

One can read this interview with the Austrian school economist Jesús Huerta de Soto in which he praises the “benefits” of deflation:
Jesús Huerta de Soto, 2014. “Deflating the Inflation Myth,” Cobden Centre, 7 December.
He makes an eyebrow raising comment here:
“[Interviewer:] Does that mean we should be happy about deflation?

[Huerta de Soto]: Certainly. It is particularly beneficial when it results from an interplay of a stable money supply and increasing productivity. A fine example is the gold standard in the 19th century. Back then, the money supply only grew by one to two percent per year. At the same time, industrial societies generated the greatest increase in prosperity in history.”
Jesús Huerta de Soto, 2014. “Deflating the Inflation Myth,” Cobden Centre, 7 December.
Really? The deflationary period of the 19th century generated “the greatest increase in prosperity in history”? What is the evidence for that?

Now it is true that there was in virtually all nations a long-run deflationary trend for most of the 19th century, even if punctuated by inflationary booms outside the 1873 to 1896 period (which was marked by almost persistent deflation).

Huerta de Soto’s statement can only mean that the gold standard era had an historically unprecedented real per capita GDP growth rate compared to all other eras both before and since.

Let us look at the average OECD real per capita GDP growth rate estimates and data for various periods over the past three centuries:
1700–1820 – 0.2%
1820–1913 – 1.2%
1919–1940 – 1.9%
1950–1973 – 4.9%
1973–1990 – 2.5%
(Davidson 1999: 22).
Uh oh.

This is what happens when an Austrian economist puts his foot in his mouth and makes statements in accordance with Austrian ideology but not backed by the empirical evidence.

As we can see, the industrial revolution of the 19th century under the gold standard most probably generated the greatest increase in real per capita wealth in history up to that time, but its record has since been surpassed. The best period of real per capita GDP growth was the 1950–1973 era: the time of Keynesian economics and modern macroeconomic management and also a much higher level of government intervention in the economy, both before and since.

But, the critic might counter-argue, didn’t Huerta de Soto really mean it was that specific deflationary period of the 19th century – which is normally taken to be the 1873 to 1896 era – that generated “the greatest increase in prosperity in history”? Possibly that is what he meant.

We can look at the data for the UK, the US and Germany below from 1873–1896 (with data from Maddison 2003):
UK Average per capita GDP Growth Rate 1873–1896: 1.057%
US Average per capita GDP Growth Rate 1873–1896: 1.422%
German Average per capita GDP Growth Rate 1873–1896: 1.495%.
Since these were the most advanced, fastest growing economies of the late 19th century, it is unlikely any other nations achieved an average per capita GDP growth rate higher than them, and certainly not in numbers large enough to affect the average for that era.

So even within the 1873 to 1896 era the figures do not seem to deviate too far from the 1820–1913 OECD average. They were also inferior to the Keynesian golden age of capitalism from 1950–1973, which remains the best period in human history for real per capita output growth.

One can also note that the deflationary era of 1873 to 1896 produced deep business pessimism at the least in the UK (and possibly other nations too) resulting from the “profit deflation” or profit squeeze of that era, and there were also protracted economic problems in the 1870s and 1890s. In the US, this period coincided with the free silver movement and bimetallist political movement that opposed the gold standard, arising in part from the debt deflationary distress to debtors in that era.

In short, the idea that under the gold standard the industrial nations “generated the greatest increase in prosperity in history” at any time before or since is untrue, and is nothing but a libertarian myth.

BIBLIOGRAPHY
Davidson, P. 1999. “Global Employment and Open Economy Macroeconomics,” in J. Deprez and J. T. Harvey (eds), Foundations of International Economics: Post Keynesian Perspectives, Routledge, London and New York. 9–34.

Maddison, Angus. 2003. The World Economy: Historical Statistics. OECD Publishing, Paris.

28 comments:

  1. I wonder how Austrian economists can advocate both the gold standard and the general cornucopianism of natural resources required for economic growth. These ideas make contradictory assumptions, because the gold standard presupposes Malthusian limits to the gold supply.

    ReplyDelete
    Replies
    1. Wage and price flexibility and deflation will miraculously cure all ills and problems -- as we all dance with magical unicorns all day long marvelling at the power of free markets.

      Delete
    2. I'd like to know if Ludwig von Mises depended on an administered wage from ad man Lawrence Fertig when Fertig bribed New York University to give Mises an office and pretend that he had a job there as a professor.

      And I keep wondering when economic chaos will ensue from the practices of the Mises Institute, which ignores price signals when it produces and gives away for free ebook versions of Mises's writings.

      Delete
  2. Given the Austrians' supposed high-regard for the wisdom of individuals, it is amusing that they have so little heed for the assessments of the people who actually experienced the 1873-1896 period. Perhaps a hundred years hence, there will be Austrians arguing that all of us who perceived a slump following 2008 were wrong too.

    ReplyDelete
  3. It's curious to me that Austrians such as de Soto hold this about deflation. I've seen others say it, too. Reisman and Selgin both come to mind, despite the latter being only loosely affiliated with the school.

    Thing is, they're getting it twisted theoretically, since general deflation only occurs under capitalism by way of crises -- such as those that occurred with some frequency in the 19th century. Thus, the grand irony: the sort of generally agreeable deflation he's describing is not capitalist at all, but socialist in character, the result of coordination instead of anarchic competition.

    Economists sure have some funny ideas about economics.

    ReplyDelete
  4. Well done, LK! But there’s another statement by de Soto in the short excerpt you provided, which we should also be at least suspicious about: “Back then, the money supply only grew by one to two percent per year”. Just how does he know that? What exactly is he measuring, and how? The “classic Gold Standard era” is circa 1875-1914, i.e. late 19th century. Before that, it was only Britain that was on a proper Gold Standard. Most other countries, if they were on anything, they were probably on a Silver Standard (yeah, I know all about Bi-Metallism, but most of the time this was really a silver standard). Further, there were significant gold discoveries in the 1850s, in California and Australia, and later in South Africa. The discoveries of the 1850s helped alleviate a little the pain of the crunching deflation in Britain, the one T. Attwood and his “Birmingham School” were bitterly complaining about (we’ve covered that, as I’m sure you remember) and may have played a role in that school’s marginalisation. There was even a slight rise of silver prices around that period which, iIrc, briefly put France on a gold-standard for a little! These discoveries (and the later ones in South Africa) must have caused the money supply to explode, at least in terms of pure gold (admittedly, it is questionable if this is how you should measure it, but if you really believe in the gold standard, you should!). So where does de Soto get his “one to two percent per year”?

    ReplyDelete
    Replies
    1. Yes, those are some good points.

      Possibly de Soto is thinking of the deflationary 1873-1896 period in which (if memory serves) there was a slower growth rate in money supply in the order of “one to two percent per year”.

      But, as we have seen, the best real per capita GDP rates in that era don't support his statement.

      Delete
    2. 1873 – 1896 won’t work either; at least not for the US. Friedman & Schwarz, A Monetary History of the United States 1867 – 1960 (1963) have a table A-1 at p. 703 et seq. entitled “Currency Held By the Public and Deposits Seasonally Adjusted, 1867 – 1960”. The total passes from 2,306 million USD in Feb. 1873 to 5,959 million USD in June 1896. According to an online calculator I found, this is a compounded annual rate of growth of 7.58 %, nowhere near the “one to two percent per year”.

      Delete
    3. That is very interesting, thanks for that data!

      Delete
    4. Digging around a bit more, I found three centuries of data on the UK economy, courtesy of the Bank of England. M0 in 1873 is said to be 139 and stands at 165 in 1896. That’s a Compound Annual Growth Rate (CAGR) of 1.74%. M4, for the same period, passes from 705 to 990; CAGR of 2.65%. There, we’re nearer to the 1 to 2 percent. However, note that the consumer price level, during the same period, is stated to have passed from 1.69 to 1.39! That’s a compound annual rate of DE-flation of 1.49% (prices dropping, on average, by 1.49% per year). Theoretically (quantity-theory-of-money theoretically), with such deflation, the money supply should have been shrinking, not growing.
      Here’s the link for the data, it directly opens an Excel spreadsheet:
      http://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&cad=rja&uact=8&ved=0CB8QFjAA&url=http%3A%2F%2Fwww.bankofengland.co.uk%2Fpublications%2FDocuments%2Fquarterlybulletin%2Fthreecenturiesofdata.xls&ei=OXiWVPfSKMj4Us2-hKgL&usg=AFQjCNHdilZGDt4Az7OCUC1dxo6LcWquzA&bvm=bv.82001339,d.d24

      Delete
    5. That is excellent work -- I will have a good look at this.

      Just one point though: the quantity theorists do have an explanation for the deflation.

      Take the standard form of the Cambridge Cash Balance Equation:

      M = kdPY

      where M = the quantity of money;
      kd = the amount of money held as cash or money balances;
      P = the general price level;
      Y = real value of the volume of all transactions entering into the value of national income (that is, goods and services).
      ---------------
      then, assuming that kd remained more or less stable in the 1873–1896 period, the quantity theorists would argue that Y was growing but M not at a sufficient rate to allow a neutral price level.

      Delete
  5. Yes, I see what you mean. Clearly, my earlier comment, that because the price level was dropping, money supply should have dropped too, is wrong. I obviously forgot that growing GDP pulls, or is supposed to pull, in the opposite direction. By the way, the same Excel spreadsheet I linked to above also has GDP estimates for the UK economy. There are various series of data for 1873 – 1896, but they all point to, roughly a CAGR of GDP for that period of 3.2 - 3.3 %. Now, if you take that result and the 1.49% drop in price level during the same period and you plug them into the Cambridge Cash Balance Equation you provided, it turns out (unless I’m doing the math wrong) that, indeed, the money supply during that period should have grown by 1.75% which almost exactly the growth of M0 during that period, indicated above. This is such a neat result that I’m wondering if the GDP estimates are not extrapolations from the money supply or something. M4 growth comes in far higher, though, at 2.65%.

    ReplyDelete
    Replies
    1. " This is such a neat result that I’m wondering if the GDP estimates are not extrapolations from the money supply or something"

      That is exactly what occurred to me!

      At any rate here are the yearly real GDP growth rates from Maddison's The World Economy: Historical Statistics (2003):

      http://socialdemocracy21stcentury.blogspot.com/2012/10/uk-real-gdp-18301918.html

      Delete
  6. Re previous comment. My calculation is probably wrong, since I looked at the CAGR of Real GDP for the period in question. But, surely, in the Cash Balance Equation, you need nominal GDP. Nominal GDP in 1873 is stated to be 1259 ("Composite estimate" column) and in 1896, 1544. We assume kd is constant (and therefore ignore it) and we multiply these amounts by the corresponding price levels (1.69/1.39) and we get, respectively, 2127.71/2.146.16, i.e. a growth of M of 18.45 for a period of 13 years, which corresponds to a CAGR of the money supply of 0.07%. In other words, with such a drop in the price level, the money supply should have remained almost constant, given the rate of growth of GDP. So my point was valid, after all, the data do not support the Cambridge Cash Balance Equation, unless you posit that kd changed, also.

    ReplyDelete
    Replies
    1. This is tricky stuff.

      For me, I think the crucial issue is the prior assumptions the quantity theory requires to be true, as follows:

      (1) the money supply is exogenously determined, and there is an independent money supply function;

      (2) the assumption of long-run money neutrality, and

      (3) the direction of causation as assumed in the quantity theory equation is from left to right (that is, from the money supply to the price level).
      ------------
      Were these true in the pre-1913 era in Britain?

      Number (2) is false, even in the 19th century.

      On (1) and (3) gold would appear to be exogenous, but were, say, countries with central banks like England (1) able to make central bank banknotes a very good substitute for gold, (2) and was the Bank of England increasing its supply of banknotes to the point where they were meeting the demand for reserves by banks and thus making the broad money supply endogenous?

      I have been wrestling with this question, but have yet to get hold of Forrest H. Capie and Alan Webber's A Monetary History of the United Kingdom, 1870–1982. Volume 1. Data, Sources, Methods (London, 1985) which, I think, has the empirical data to answer this question.

      Delete
    2. If you want the data from Capie & Webber, then go to the spreadsheet I linked to, it’s got them. In fact, now that I check it again, I see that the data on M0 and M4 I quoted above come from “Capie and Webber (1985) and Bank of England/ONS”. So, whatever it is you’re looking for, you’ve found it!

      Delete
    3. Thanks -- that excel data is brilliant!

      Delete
  7. "It is particularly beneficial when it results from an interplay of a stable money supply and increasing productivity"

    I think what he means is stable nominal income, not stable money supply. Stable money supply means nothing if nominal income is collapsing.

    ReplyDelete
  8. LK,

    "the 1873 to 1896 period (which was marked by almost persistent deflation)"

    According to stats you present elsewhere, during that 23 year period there were 9 years without deflation. So for about 39% of that period there was no deflation.

    http://socialdemocracy21stcentury.blogspot.co.uk/2012/10/the-gold-standard-did-not-prevent-price.html

    ReplyDelete
    Replies
    1. Yes, there was apparently a 0% inflation rate or something close to it in some years and brief inflation year in 1887.

      So, yes, there was not a persistent continuous deflationary trend in the period in question.

      This clearly does not help Huerta de Soto's case.

      Delete
  9. LK, what are Maddison's growth figures for the 1950-1973, 1973-1990 periods?

    ReplyDelete
    Replies
    1. I am fairly sure the ones for the OECD are taken from Maddison:

      1950–1973 – 4.9%
      1973–1990 – 2.5%
      (Davidson 1999: 22).

      Europe had higher figures than the US, and my calculation of the US data from Maddison gives its average per capita growth rate for 1948–1973 as 2.30%.

      Delete
  10. LK, I've just been looking back at some of your old posts on the late 19th century.

    This post indicates that there were 11 years between 1873-1896 in which there were recessions (not necessarily lasting a whole year though):

    http://socialdemocracy21stcentury.blogspot.co.uk/2013/02/19th-century-deflation-and-recession-in.html

    yet other posts suggest that there were far fewer recessions over that period:

    http://socialdemocracy21stcentury.blogspot.co.uk/2012/08/davis-on-us-recessions-in-19th-century.html

    This is a bit confusing. What explains this difference?

    ReplyDelete
    Replies
    1. (1) http://socialdemocracy21stcentury.blogspot.co.uk/2013/02/19th-century-deflation-and-recession-in.html

      This post shows deflation and inflation, but not every year of deflation was a year of recession.

      (2) http://socialdemocracy21stcentury.blogspot.co.uk/2012/08/davis-on-us-recessions-in-19th-century.html

      This post shows Davis's list of postulated recessions on the basis of his industrial output index.

      But it is likely that his list is not complete, and sometimes it disagrees with the list of recessions in Balke and Gordon (who make new calculations of US real GDP).

      Delete
    2. "This post shows deflation and inflation, but not every year of deflation was a year of recession."

      In that post you write:

      "In yellow, I have highlighted the periods where price deflation is correlated with recessions identified by Davis"

      and there are 11 years highlighted in yellow between 1873 and 1896, which indicates 11 11 years between 1873-1896 in which there were recessions correlated with deflation.

      The second post, however, indicates that there were only four years over that period in which there were recessions.

      I'm wondering how you reached the figure in the first post.

      Delete
    3. Hmmm ... here is the data from Davis in the second post:

      US Recessions in the 19th Century
      Years (Peak–Trough) | Recession Length (years)
      1873–1875 | less than 3
      1883–1885 | 1
      1892–1894 |
      1895–1896 |
      (Davis 2006: 106).
      ---------
      Is it that you are looking the "Recession Length " column? Of course I have not put in the lengths for all of them (following Davis) because there is some doubt about the precise lengths of some.

      However, there were recessions in 1892–1894 and 1895–1896. Of course it is true that 1894-1895 had some kind of recovery in Davis, but it is hard to indicate this in my data, so I left 1894 and 1895 in yellow.

      Looking back at Davis I see he lists a recession from 1883–1885 but strangely only lists the years as "1" -- which in reflection looks like a mistake.

      Anyway, if one wants a rough year count I count about 10/11 years of recession in Davis's data.

      Delete
    4. ok, sorry I misunderstood your second post. Thanks for clarifying.

      Delete