Preface (always great if you’re starting a blog post with one of these…)
Amidst this current time of a pandemic causing unemployment at never before seen levels, my interest in the last time America had an unemployment crisis of this magnitude, the Great Depression, was re-ignited to help give a sense of proportion to the modern crisis.
Also, Noah Smith wrote a column in Bloomberg focused on the New Deal that totally didn’t trigger me, make me go, acshuually, I know more than him about this, and caused me to spend a few hours pouring over some of my old notes and writings from economics grad school regarding the Great Depression to prove him wrong.
Anyway, as a result of all of this, the following is significantly higher effort and more wonkish/academic than my average post will be (edit 3/24/2021 – turns out all of my posts are high effort and wonkish/academic and I am reserving the more normie stuff for my podcast). If you are a normal person, I would recommend beginning with my explanation towards the end with the heading for people that don’t find economics sexy, then review the pictures, and read the remainder of the post if you are still interested.
The next few sections assume some intermediate knowledge of macroeconomics, but I also hope to be more informative with loads of high-end sources and is less in need of correction.
The Great Depression and its Causes
From August 1929 to March 1933, the United States suffered the largest economic contraction in its modern history, with NBER estimating the 43 months of contraction were virtually twice the length of any other recession since 1882. Shockingly, this economic contraction occurred in a decade where the U.S. experienced its largest total factor productivity growth in the twentieth century (Field 2006, p.1). It is with this in mind, Ben Bernanke writes, “To understand the Great Depression is the Holy Grail of macroeconomics.” Understanding the roots of its causes and the exact nature of the ensuing 1933 to 1940 recovery (although as I later discuss, this is not a period immune from economic contraction) is critical for preventing similar downturns, such as the recent financial crisis.
A tight, deflationary monetary policy was at the epicenter of what would become the Great Depression. Irwin (2012, p.1) suggests nowhere was there more fault than the United States and France because they sucked…the world dry of its gold reserves. He notes the immense deflationary pressures experienced in other nations, when the two countries held 60 percent of the world’s total monetary gold stock by 1932. France in particular is so disproportionately to blame, that he writes, “Even Milton Friedman later said that, had he been fully aware of France’s policy, he would have revised his view on the origins of the Great Depression” (p. 5). (If interested, a beginner’s guide to the gold standard and its implications occurs in the final section of this post).
Eichengreen and Temin (2010, p. 437-438) add more to Irwin’s cause. They argue the “gold standard mentalité,” a prioritization of the gold standard over employment or output stabilization, among leading industrialized nations was to blame for the crisis. Central banks among the aforementioned nations in particular made an unexceptional downturn exceptional by hysterically raising interest rates to purchase even greater sums of gold. These higher interest rates in turn led to a banking crisis, as they destabilized and closed en masse, which in turn caused even greater compounding negative shocks to production, prices and employment. Bernanke (1983, p. 269) agrees, though he ultimately only finds monetary factors to be responsible for little more than half at most of the 1930 to 1933 downturn.
Although Bernanke (1995, p.2) believes monetary factors to be the sole largest cause of the downturn, he instead offers two main nuanced origins of the Great Depression: a financial crisis that was induced by the deflationary monetary policy, and real-wages that were maintained at levels greater than the market-clearing rate. The former of which is caused by an increase in the cost of lending credit, which developed into a ‘credit crunch’ of more conservative lending behaviors among banks and great, long-lasting negative shocks to aggregate demand (Bernanke 1983, p. 257). Jalil (2014, p. 259) furthers Bernanke’s claim, noting banks under jurisdiction of the heavily monetary interventionist Atlanta Fed were less likely to fail than others under bordering regional Fed banks, which suggests panics were induced by illiquidity – not inherent instability in the banks themselves. However, Bernanke notes the latter concept of artificially high real-wages is difficult to reconcile with the concept of often-assumed economic rationality. Thankfully, Jonathan Rose and Lee Ohanian offer such an explanation.
Ohanian (2009, p. 2310-2311) develops a theory that President Herbert Hoover’s pleading with industry and labor to keep wages high in face of the recession in exchange for not unionizing helps to explain part of the “persistent and large labor market failure” of the Depression. Rose (2010, p. 843) offers some additional empirical evidence of this theory, noting persistently high wages paid by industrialists who had met with Hoover on Nov. 21, 1929.
Lastly, Christina Romer (1990, p. 597) argues that the collapse in stock prices during what is often used to describe the onset of the Great Depression in 1929 caused uncertainty as evidenced by that expressed by economic forecasters and individuals purchasing greater durable goods. The ensuing decline in consumer spending as advocated by Romer offers a more Old Keynesian explanation of potential causes of arguably the worst economic crisis in American, if not world, history.
Recovery, the New Deal’s Impact, and the Return of the Depression
There is little doubt that Franklin Roosevelt played a starring role in the Great Depression, though whether his actions were overall positive, negative or a mixed bag remains much more controversial. Temin and Wigmore (1990) contend the Roosevelt regime by offering experimentation and dollar devaluation was instrumental in ending the crisis. They write his “championing the virtues of inflation” in particular positively shifted expectations in March 1933, his inauguration, and resulted in greater investment, nominal farm prices and incomes.
It should be noted, the growth experienced between March and July 1933 are “by far the largest” of any four-month period dating back to 1884 (Taylor and Neumann 2016). Eggertsson (2008) furthered this idea of Roosevelt’s regime change. He uses a DSGE model to discover Roosevelt’s monetary policy of ending the old gold standard dogma, as well as his expressed desire to increase government spending to (compared to the past) large peacetime deficit levels, shifted expectations outward and drove the recovery. Additionally, Romer (1992) found aggregate-demand boosted by a large influx of gold and ensuing monetary stimulus, not market self-correction, was the cause of almost all of the recovery during the Depression. However, neither Romer nor Eggertsson seriously explores why one of the periods of immense growth was followed by an abrupt downturn. They share the similar mistake of treating the recovery period as essentially a singular entity.
Although they agree inflation expectations changed as a result of the Roosevelt administration and this explains much of the recovery, Taylor and Neumann (2016) note other causes of recovery, such as greater consumer confidence. In other words, the market was already self-correcting to an extent, irrespective of any one of Roosevelt’s policies per se. Additionally, some of Roosevelt’s policies were more limited than appears at first glance. For example, even though some evidence suggests New Deal spending was directed to areas with relatively high unemployment because of the recession, it is likely the spending was more strongly correlated to variables of political clout, such as an area having a representative on a Congressional appropriations committee (Anderson and Tollison 1991).
Furthermore, although work relief during the First New Deal at the height of the Depression likely increased private employment and earnings, that of the Second New Deal likely resulted in private crowding out and decreased employment (Neumann, Fishback, and Kantor 2010). Meanwhile, Roosevelt’s high wage policies likely had a detrimental effect on the recovery in general.
The most controversial policy of Roosevelt’s was the National Industrial Recovery Act, especially its creation of the National Recovery Administration. The NRA was designed to cease “cut throat competition” that was believed to be harming the economy and to give workers more money. Naturally, the way to do this was by suspending antitrust law and essentially forcing most major industries to cartelize themselves in arrangements that varied by industry, but generally included production quotas, fixed prices, set minimum wages for workers and limited the maximum hours one could work. There was also a massive propaganda push to only shop at NRA approved shops and to view firms that did not willingly participate as traitors.
Taylor and Neumann (2016) blame this cartelization and the artificially high wages introduced by Roosevelt under the National Industrial Recovery Act (NIRA) as primary causes of the contraction, as well as the act’s eventual unconstitutionality and repeal being major factors in the eventual continuing recovery.
Cole and Ohanian (2004) go so far as to argue the recovery was never really there at all, but instead was weak as evidenced by GNP, consumption, investment, and private hours-worked being well below their 1929 trend for almost all of the 1930s. However, they construct a dynamic general equilibrium model to determine that because of the NIRA and artificially high wages (especially in manufacturing), the recovery was strangled due to a negative supply shock. Though he does not go so far as Cole and Ohanian, Sumner (2015) similarly argues in The Midas Paradox that the devaluation of the dollar had largely solved the aggregate demand problem by 1934, but drastic labor market interventions such as the NIRA caused a leftwards shift in aggregate supply and thus ensuring the second recession.
John Maynard Keynes said it best himself in his open letter to FDR, when he wrote:
That is my first reflection–that N.I.R.A., which is essentially Reform and probably impedes Recovery, has been put across too hastily, in the false guise of being part of the technique of Recovery. . . .
I do not mean to impugn the social justice and social expediency of the redistribution of incomes aimed at by N.I.R.A. and by the various schemes for agricultural restriction. The latter, in particular, I should strongly support in principle. But too much emphasis on the remedial value of a higher price-level as an object in itself may lead to serious misapprehension as to the part which prices can play in the technique of recovery. The stimulation of output by increasing aggregate purchasing power is the right way to get prices up; and not the other way round.
In sum, one has to argue that Roosevelt’s policies were very much a mixed bag, though their net good or harm is still contestable. With that being said, I believe the NIRA was unequivocally a significantly net negative harm – the abrupt halt to recovery in 1933 is inexplicable without that being the case. Meanwhile, the public work programs and relief spending in the prolonged recession were almost certainly net benefits to the economy and most workers, albeit tremendously overrated by conventional wisdom relative to the impact of Roosevelt’s inflationary monetary policy – which is nowhere near as visible and much more arcane to the untrained eye.
For the People That Don’t Find Economics Sexy
Essentially, the gold standard caused the Great Depression. Specifically, France in particular hoarding gold. It was also abandonment of the gold standard that saved most of the world and countries not on the gold standard (such as China, which had a silver standard) managed to largely avoid the Depression in its entirety.
Under the gold standard, each country would provide a pegged exchange rate of their currency for a redeemable amount of gold at some set price. For example, let’s say the price of an ounce of gold is 10 American dollars, while it is 8 French francs. Therefore, $10 USD and 8 French francs are equivalent. To engage in international trade, countries would therefore either use the gold itself or gold-pegged currency (so that more gold could be free in circulation) for exchange, which under perfect circumstances, is fine. However, in the real world problems occur when countries over or undervalue their currency.
If a country’s currency is overvalued relative to others’ (the pegged price is too high), then their exports will drop because their goods are essentially more expensive. If a currency is undervalued (the pegged price is too low), their exports will be higher because their goods are essentially cheaper.
Knowing this, if a country wants to offset what is effectively a change in price of their exports, they would have to use monetary policy (i.e., change the money supply in their economy) to restore parity. In other words, if your currency is undervalued and therefore your exports are cheaper, a country would need to increase their money supply through an action such as changing the pegged value of their currency to offset this difference.
To explain in plainer English and a bit more simplified, let’s use the earlier example of USD and francs. Let’s say that although $10 USD and 8 francs are pegged as equivalent (because both have been determined by the country to be worth 1 oz of gold), in reality, it should be $10 USD and 12 francs that are equivalent. Therefore, when Americans are buying goods from France, they are effectively receiving a 1/3 discount because they are buying 12 francs worth of goods for the equivalent price of only 8 francs. Americans will then continue to buy French goods, and if the French do not ever change their money supply via their pegged rate, the end result is their country hoarding the world’s gold like they are some dragon.
If it seems far-fetched, this is precisely what happened in history! Despite much of the world relying on the gold standard for international exchange, France refused to change the peg of their undervalued currency, and without any inflation, managed to single handedly suck up more than 1/4 of the entire world’s gold! What are the consequences of every other country suddenly faced with a lot less money in circulation?
So here is a fairly standard supply and demand graphic for an economy, and if you’re an economist, the only thing you may not recognize is SRS, which should really say SRAS, but shitty MS paint graphs don’t just make themselves. Khan economy explains all of these concepts more here, but I’m going to explain the image as well because my first ever post wasn’t long enough already.
LRAS means long-run aggregate supply and it is vertical because ultimately what an economy can make is bounded by its factors of production, i.e., the amount of labor, land, capital, human skills, etc. These are not ultimately determined by things like the price level or how much money is circulating in an economy.
SRAS means short-run aggregate supply, and it is sloped upwards because economics is self-contradictory, this is the only supply curve that matters, and all those things that I said about the economy not being determined by prices are false because of course firms want to sell more things for more money and don’t you know the entire field of economics is all irrelevant because of the inevitable heat death of the universe anyway?
AD means aggregate demand, and you can think of it as a measure of the demand for all goods and services in an economy as expressed by the total money spent. When things cost less money, people want to buy more, hence the downwards slope.
Now, when you remove a bunch of money from the economy because the French Ancalagon the Black is hoarding all the gold, one would see a shift in aggregate demand from ADold to ADnew. Nothing has changed with the nature of the long run economy itself, there’s still the same amount of people and things, but because there’s less money circulating, the price level is lower. I.e., there is nothing inherently changed with the overall economy, everything should just cost less. There is one small problem with that though…
PEOPLE HATE TO BE PAID LESS. Good luck trying to explain to someone that actually, they should be fine taking a 5% pay cut because the French Smaug has stolen 10% of your village’s gold, so really, it’s not even a pay cut, but a pay raise. Even though there is nothing inherently wrong with the economy itself (such as say, a rampaging epidemic…), the end result is with employers’ hesitance to lower wages and workers’ (perfectly reasonable!) refusal to accept any lower wages, without enough cash to go around employers have to start firing people.
This was exacerbated by tremendous pressure and laws from the government such as the NIRA to not only minimize paycuts or hold wages constant, but to significantly raise wages. When the inflation rate is -10% like it was for 1932, this is a particularly horrible idea – even if you think there is a large role for the government to play by stepping in and giving people more money directly (which I happen to believe)!
Therefore, not only is there this contraction in employment, but banks will also have less money (i.e. gold) in their reserves, and people start hearing about this, so they start to head to banks to withdraw their money out of a sense of caution, or perhaps they had been laid off and need to use whatever savings they have. More and more people start to do this and suddenly, an otherwise perfectly solvent bank goes bust because it does not have enough cash on hand to give to all of these people at once. Not only that, but all of the layoffs and banks going bust at around the same time causes further panic that results in a vicious cycle of them both happening with greater and greater frequency. The output gap where the economy is currently producing and where it should be producing becomes gargantuan from the mass panic, unemployment, and bank failures (this is the difference between the intersection of ADold & SRAS, and ADnew & SRAS).
This is why the gold standard was so horrible and why the world abandoning it or any other metallic standard to print a bunch more money was essential to recovery. As per the chart below, even though they were remarkably stable at preserving long run price stability, this came at the cost of regular wild short-run price fluctuations. Those fluctuations caused multiple severe downwards deflationary pressures, which caused multiple recessions and is problematic for all of the aforementioned reasons.
This is also why economists in general are much less worried about inflation relative to deflation. A constant, stable inflationary increase in the money supply regardless of what is happening in the rest of the macroeconomy is THE key lesson learned from the Great Depression, and why it took more than a century and a pandemic for the US to return to comparable levels of unemployment.
It is also why the Federal Reserve is doing a noble job injecting unprecedented money into the American economy at the moment, and more importantly, the origin for this glorious meme.
