Deflations Not Easy And Benign

This article is a preliminary unedited draft section of my inflation manuscript. There are therefore references to other “sections” and charts in those sections that the reader is unaware of. That manuscript has been on the back burner for some years now. I was not happy with the text, and rather than just push it out, I decided to wait and move onto other projects. I have done a partial re-write and update, and I am now happier with the text. No promises as to when it hits bookstores.

This section discusses an idea that pops up in various contexts. The basic premise is a reaction against the idea that “deflation is bad,” which is a common view that was cemented by The Great Depression of the 1930s (and not the earlier Great Depressions, which I will get back to later in this section).

For reasons that will become clear as we discuss the topic further, the idea that deflation is “natural” is an outgrowth of Gold Standard thinking and is often pushed by people who wish to return to the Gold Standard (e.g., Austrian economists). As is typical for Austrian thinking, the ideas show up in financial/economic commentary written by people who are repeating ideas that have echoed around Austrian economic circles for about a century now, and any scholarly sources have long been lost track of. I will outline the issues with the basic premise but then turn to a more reliable source (which is less pro-deflation than pop Austrians).

Deflation and the Great Depression

The figure above shows the long-run change in the inflation index for the U.K. starting in 1900 (the second half of the figure in Section 1.3). The focus in this discussion is the period ahead of World War II (1939). There was an inflation spike due to World War I (1914-1918) that ended in 1920, which then reversed (rapidly, then slower). The First World War was a total war, and inflation was one way in which the economy absorbed the demand for wartime needs. The U.K. de-pegged its currency from gold to allow wartime finance to work (a standard policy step). The deflation was the result of the decision of the British government to return to the pre-war gold parity (exchange rate of the pound versus gold). My preferred reference for this period is Barry Eichengreen’s Golden Fetters: The Gold Standard and the Great Depression 1919-1939. The book gives a global history of this period, with the argument that the return to old gold parities were the driving force in the Great Depression (the 1930s). Although the U.S. economy boomed in the 1920s, this was less true for the United Kingdom courtesy of the problems caused by the austerity policies needed to return to the old gold parity.

To briefly explain why restoring the old parity was economically – and politically – painful, we need to look at the mechanisms behind the Gold Standard. To credibly promise that British pounds could be exchanged for gold at a certain exchange rate, Britain needed gold reserves at a certain coverage ratio relative to its monetary base. The situation is slightly more complicated by the fact that the exchange rate versus the U.S. dollar was operationally more important than gold itself – the U.S. dollar was the hardest currency in the system as they had received large gold inflows during World War I as it remained neutral until 1917, and so had large gold inflows as they paid for war materiel. To generate the gold needed for the gold cover, the government had to raise interest rates (to attract gold inflows) and run tight fiscal policy (cutting spending, raising taxes) – or hope that the Americans loosened fiscal policy. As such, the problem was not the deflation itself, rather the need to throttle growth with fiscal and monetary policy.

The post-World War I deflation was the last great set of deflations in the developed world. Although some economists continue to argue about the causes of the Great Depression of the 1930s, the fact that it occurred alongside deflation has certainly made it easier to describe deflation as being bad for the economy. This historical experience was also supplemented by mainstream economic theories that suggested that deflation would generate economic weakness that cannot be solved via monetary policy – the preferred business cycle management tool for the mainstream. The “liquidity trap” was the popular name for the problems alleged to be created by deflation. The policy rate allegedly could not go negative (it turns out that this was not true), and so it would not be possible to create the negative “real rates” (the nominal exchange rate less the inflation rate) that allegedly is needed to break out of a depression. As my wording suggests, I am not a fan of the “liquidity trap” theory, but it did represent the mainstream consensus at one point.

Post-World War II: Deflations are Mild

Once we get past World War II, the rise of Keynesian economics and the effectively looser Bretton Woods system meant that aggregate deflation was a rarity. (Under Bretton Woods, other countries pegged their currencies to the U.S. dollar, and the U.S. in turn had a gold peg. However, the U.S. ended World Wars One and Two with so much of the world’s monetary gold that the gold constraint was fairly loose until the 1960s. President Nixon ended gold convertibility once the policy became too painful. But until that happened, countries were far less constrained by gold availability than during the classical Gold Standard years.) In the major developed countries, the euro periphery after the Financial Crisis was the only place to see much in the way of deflation (the price index for Greece is in Section 1.3). Like in the 1930s, that was in the context of a major decline in nominal GDP. Although Japan was allegedly in the thrall of “deflation” after the mid-1990s, the figure in Section 1.3 shows that this was really a period of price level stability.

The absence of deflations in any time series history people normally looked at made it harder to argue that deflation was a “normal state of affairs.” Instead, people arguing that “deflation is natural” grabbed sub-components of the CPI. They often grabbed price histories particular goods (like television prices), but we can just look at goods versus services to see the big picture.

The figure above shows the post-1990 experience for the level of two major categories of the CPI in the United States – Services and Commodities (goods). Services price rises outstripped those for goods, although there was a little bit of a catch-up in goods prices after the pandemic. Note that “goods” is an extremely wide category and will include things like gasoline that face resource depletion, and so we do not have much outright deflation. But if we drill into some sub-categories of manufactured items, particularly electronics, deflation was the norm (until 2020, at least). The case of falling computer prices was covered in Section 3.3, but televisions are also a popular category for deflation anecdotes.

The logical leap then taken is saying that this demonstrates that falling prices are “natural”: if television and computer prices fall over time, why not everything else? This is unfortunately a very large leap – electronics goods fall because there are countless busy beaver engineers creating new technologies and optimising existing technologies. This constant improvement in production capabilities is normally described as “productivity growth.” (The situation is somewhat complicated by international trade. Part of the deflationary pressures in the developed countries after 1990 was the result of China joining the international trading order, and the Chinese leadership exuberantly following the export-driven growth models previously followed by Germany and Japan and then East Asia after World Wat II. The downward price pressure partly reflected lower wages then greater productivity growth due to the advantages of economies of scale. So “productivity” here could be viewed as a more generic term than what is normally referred to in the economic theoretical literature) By contrast, engineers cannot greatly reduce the costs of managing an apartment block, so productivity growth in rentals is negligible. Although there are exceptions, the explanation of the previous chart is that productivity growth in manufacturing is higher than productivity growth in the service sector.

Furthermore, it is clear that societies are concerned about the aggregate price level, not just particular goods and services. What we saw in most of the developed countries from 1995-2020 was that inflation rates typically stuck around 2% (or 0% in Japan). Policy interventions would be made if inflation gets too high or low. So, it does not matter that goods prices tended to deflate – what mattered was overall inflation (around 2% in the United States). Even if we lowered the overall inflation rate, the productivity disparity still exists, and goods prices would have fallen relative to the overall price index, and services faster.

In order to get to premise that “deflation is natural,” we need to accept that “societies targeting deflation is natural.” That is exactly the intent of most of the people pushing the “deflation is natural” story – they want to return to a policy framework like the Gold Standard, where deflation was a typical outcome.

Gold Standard Experience: 1800-1900

The figure above is the first half of the long-run price index series from the Office of National Statistics for the United Kingdom, running from 1800-1900. (Since the United Kingdom was the linchpin of the global economy in that era, this price experience matters more than would the case of countries at the periphery of the system – like the United States.) If we look at the entire century, the price index fell by a compounded annual rate of about -0.4% per year. However, most of the volatility of the series was during 1800-1820, in which the Napoleonic Wars were fought (“Oceans are now battlefields”). Britain was in the now-familiar position of fighting wars against a continental power with a much larger army, and so it relied on its command of the sea. The exertions of the wars caused an inflation, and then the return to normality resulted in deflation as the United Kingdom restored its earlier gold parity.

The price level was largely stable until the mid-1870s, at which point the “Long Depression” hit. This was a worldwide event, and the dating of the recessions depended upon the country. For the United Kingdom, some sources use 1873-1896. It was referred to as “The Great Depression” at the time, but the name was revised to “The Long Depression” when the Great Depression of the 1930s arrived.

The Long Depression was more uneven in its impact than the Great Depression (1930s). The agricultural sector was hit hard, but manufacturing was absorbing new technologies and expanded. As a result, the contractions in real GDP appear to be muted – although estimates of real GDP are research series produced by academics, and one might have concerns with the methodologies.

With that context in mind, it is less clear that the -0.4% annual deflation rate is meaningful as a reading of a “normal” economy – it reflected the post-war deflation and then the results of The Long Depression. To the extent that one feels that The Long Depression was just the story of the more productive manufacturing sector pushing the less productive agricultural sector out of the way, the resulting deflation can be termed “good” (if you were not a farmer). The article by Bordo, Lane, and Redish in the references argues that this period can be seen as a period of “good deflation.” However, it is very hard to see how a key period of the Industrial Revolution as being an environment that we expect to be sustained and thus “natural.” Furthermore, banking crises were common. An article by Kenny, Lennard, and Turner in the References identifies 1772, 1815-6, 1825-6, 1841, 1866 and 1929-30 as being banking crises, while 1847, 1857, 1878, 1890 and 1914 were commonly cited dates for banking crises yet did not have large numbers of bank failures in their data set. Some free-market fans view periodic banking crises as being a good thing as it purges the system, but there is not a lot of evidence that the broad public agrees with that stance.

Concluding Remarks

Whether or not one feels “deflation is natural” probably depends on one’s political economy leanings.

• If you believe that the Gold Standard (or similar system involving cryptocurrencies) is “natural,” then periods of deflation are to be expected. However, it is unclear to what extent that the deflation of The Long Depression would repeat in a more stable technological/economic environment.

• If you believe that the government tying its fiscal and monetary policy to a price control scheme for a collectible is a violation of the premises of democratic governance and free markets, you then end up with free-floating currencies. In that environment, the revealed preference is for low and stable inflation rates (possibly near 0%, as seen in Japan and Switzerland).

The political problem for the Gold Standard is that the reactionaries who support it have not kept up with the political trends since 1914. The advent of total wars that drew in almost the entire adult population into the war effort meant that anything other than a universal voting franchise untenable. Sacrificing the economy on a cross of gold to benefit a handful of gold holders is not a politically sustainable policy.

References

• Eichengreen. Barry Golden fetters: the gold standard and the Great Depression, 1919-1939. Oxford University Press, 1992.

• Krugman, Paul R., Kathryn M. Dominquez, and Kenneth Rogoff. “It’s baaack: Japan’s slump and the return of the liquidity trap.” Brookings papers on economic activity 1998.2 (1998): 137-205. The popular writings of Paul Krugman at the New York Times would be a readable introduction to the ideas of the liquidity trap, but finding them may be difficult as search engine functionality decays.

• Bordo, Michael D., John Landon-Lane, and Angela Redish. “Good versus bad deflation: lessons from the gold standard era.” (2004). This paper looks at aggregated economic statistics, and my concern is that this is missing the trauma in the agricultural and banking sectors.

• For the history of banking crisis dates, I referred to the dates in this NIESR blog post: https://niesr.ac.uk/blog/history-uk-banking-crises-time-different
  The blog post was based on data in this paper: Kenny, Seán, Jason Lennard, and John D. Turner. “The macroeconomic effects of banking crises: Evidence from the United Kingdom, 1750–1938.” Explorations in Economic History 79 (2021).

Email subscription: Go to https://bondeconomics.substack.com/ 

(c) Brian Romanchuk 2026