Saturday, November 15, 2008

The Oil Shock of the 1930s: Another Factor?

The chart above shows the real monthly price of oil from January 1931 to December 1939, using monthly oil prices adjusted for inflation using monthly CPI data, both data series from Global Financial Data (paid subscription required). Between 1931 and 1934, real oil prices more than doubled from $20 to $40 (in 2008 dollars), and peaked at $42.59 by the summer of 1937. The oil shock of the early 1930s was roughly equivalent to the oil shock that resulted in a doubling of oil prices between 1979 and 1981, contributing to an 18-month recession and 10.8% unemployment.

Along with contractionary fiscal policy via tax hikes, contractionary monetary policy, and huge increases in tariffs and trade protection, perhaps it was also the "energy shock of the 1930s" that helped turned what would have been a fairly ordinary recession into the Great Depression?


At 11/15/2008 12:54 PM, Blogger spencer said...

British Petroleum reports very different data for real oil prices in the 1930s.

according to them it was:
(2007 $)


BP has it essentially unchanged from 1929 to 1939 as compared to your more than doubling.

At 11/15/2008 1:07 PM, Blogger Mark J. Perry said...

spencer: 1) I'm showing monthly data 1931 to 1939, the BP data is annual data (over a different and longer time period), which would smooth out the highs and lows during the year. 2) The real price of oil according to BP almost doubled from 1931 to 1934, and did approximately double between 1931 and 1937.

At 11/15/2008 1:54 PM, Anonymous Fred said...

A great day to be honored by the snark of Angry Bear's own Spencer.

Or is it just that all of Angry Bear's anonymous flying monkeys were busy with other flying monkey stuff?

At 11/15/2008 2:08 PM, Blogger the buggy professor said...

An interesting post, Mark. First time I ever heard about this oil-ride rise.

1) How significant the doubling of oil might have been in the early 1930s would probably hinge in large part on how much was used in transportation in that period, compared to the current decade.

In turn, its level of use in transportation would no doubt depend on the number of cars per American as opposed to today's number.


2) So, if somebody has time --- back here in Santa Barbara my wife, my dog, and I myself have had to evacuate our house in the hills of Santa Barbara, not more than a mile from the raging fire in the steep foothills to the north and west --- if somebody has time, he or she might run a google search for the number of cars or cars per capita and see what's what.


3) I'm pretty sure that, additionally, coal would have been much more extensively used in heating systems than it is now, what with hydroelectric power and nuclear power, with a tad here and there of wind-generated power and solar systems.

And of course air conditioning hardly existed as a demanding energy-usage.

Michael Gordon, AKA, the buggy professor

At 11/15/2008 2:46 PM, Anonymous qt said...

Buggy Prof,

Would also have to consider industrial and agricultural usage. Additionally, oil is used today to create a wide variety of petroleum based products which would not have existed in the 1930s.

Transportation accounts for 2/3rds of U.S. oil consumption today

At 11/15/2008 3:14 PM, Anonymous Anonymous said...

Why does monthly data compared to annual data make such a huge difference? It shouldn't.

The BP data is the black gold standard.

At 11/15/2008 3:39 PM, Anonymous Anonymous said...

Follow up: The Global Financial Data (GFD)data set is suspect. Here is a 2004 article from GFD. There is a chart in the article indicating that in 2004$ the price of oil was $11.13 in 1930 and $13.71 in 1940.

Something is screwy with the GFD data charted by Carpe Diem.

At 11/15/2008 3:40 PM, Blogger Mark J. Perry said...

For 1933, the average price was $30 (in 2008 dollars), but the low was $22.59 in May, and the high was $39.32 in December. Using the average annual price of $30 misses the low of $22.59 and the high of $39.32.

At 11/15/2008 3:45 PM, Blogger Mark J. Perry said...

Description of Global Financial Data's oil prices:

"Data from 1898 to 1912 are taken from the NBER. Prices are average monthly price per barrel at wells, pipeline certificates, January 1890 to January 1895, Pennsylvania Crude at wells, February 1895-1935. Quotations are from Stowell’s Petroleum Reporter (January 1890-September 1901) and from reports by Miss Belle Hill of the United States Geological Survey of 1902. 1903 monthly quotation oil city derricks, and for 1904-1912, first of each month quotation for Oil City Derrick.' Prices from 1898 to 1941 are the price per barrel for oil in Oklahoma City."

At 11/15/2008 4:06 PM, Anonymous Michael Smith said...

In addition to the steep increase in the individual income tax rates, there were several other tax increases:

1) Hoover initiated new excise taxes on cars, movie tickets, radios, phonographs, telephone calls, telegrams, cosmetics, cameras, bank checks, stock transfers, yachts, jewelry furs and gasoline. When he became President, Roosevelt kept these taxes in place -- and in fact increased some of them.

Excise tax collections went from $540 million in 1929 to $1.364 billion by 1935. So it wasn’t merely the upper income earners that got hit with taxes. It was virtually everyone, to one degree or another. By 1935, these excise taxes exceeded the taxes collected from the individual income tax and the business income tax.

2) Both Hoover and Roosevelt raised the business income tax. In 1929, the first 3,000 of business profit was not taxed at all; anything above that was taxed at 11%. Hoover changed this to 12%. In 1932, Roosevelt eliminated the $3000 “exemption” and all business income was taxed at 13.75%. In 1936, the rate on amounts over $40,000 was upped to 15%.

3) In 1936, Roosevelt came up with the “Undistributed Profits” tax. Profits, of course, were already being taxed, as were dividends. However, Roosevelt was convinced that the reason more businesses weren’t paying their profits out as dividends or executive compensation was purely to avoid having them taxed again. (No, Roosevelt didn’t know much about the practice of retaining earnings to finance future operations or expansion.)

So Roosevelt came up with the “Undistributed Profits” tax (UP tax.) Under this plan, any profits retained by the business were subject to an additional tax on a sliding scale depending on how much was retained.

For instance, at the low end, if a company kept 1% of their net income, 10% of that amount would be taxed under the UP Tax. At the upper end, if a company kept 70% of their net income, that income would be taxed at a rate of 73.91%.

All efforts to convince Roosevelt that tax increases would not help unemployment failed.

4) Roosevelt’s Agricultural Adjustment Act started in May 1933. Its primary thrust was an effort to raise commodity prices by paying farmers to leave the land barren. To pay for this, new taxes were placed on those businesses that processed what the farmers produced. I can’t find any direct information to indicate how large those taxes were except that by the middle of 1935, Time Magazine was reporting that some 10 million dollars in taxes had been extracted from these processors. See:,9171,748845-1,00.html

5) And then, of course, there was the Social Security tax which hit both individuals and businesses.

But the insanity of Hoover and Roosevelt’s tax policies is easily matched by the insanity of Roosevelt’s NRA, which mandated price fixing among industry groups to establish minimum prices for products and services -- minimums which caused many businesses to have to raise their prices. Those that refused to raise their prices were jailed.

The NRA also mandated minimum wages for each industry. And again, those that refused to raise their wages were simply jailed.

So in a period of mass unemployment, with falling revenues and spreading business failures - Roosevelt forced both wages and prices upward.

All of this is in Burton Folsom’s book, “New Deal or Raw Deal?”.

At 11/15/2008 5:40 PM, Anonymous qt said...


The myth of FDR continues to persist despite extensive documentation of the effects of FDR's many misguided policies on labour, taxes, price setting,etc. Strange that the present crisis is invoking a renewed embrace of government intervention.

Will look for that book. Thanks for the recommendation.

At 11/15/2008 8:56 PM, Anonymous qt said...

Marginal tax rates during the Great Depression..linked article makes interesting reading particularly as we consider how to solve the present crisis.

At 11/15/2008 9:02 PM, Blogger Mark J. Perry said...

qt: That was my actually my chart from this CD post that Mankiw used, but he gave Alex Tabarrok credit by mistake!

At 11/16/2008 4:33 PM, Blogger spencer said...

Again, using the same data source as you, the swings in the real price of gas were driven more by the changes in the general price level than in nominal gas prices.

From 1929 to 1939 the nominal price of gas did not change that much.

As a matter of fact it actually fell about 20% in 1930 & 1931 and ended the decade below where it stated in 1929.

nominal gasoline prices

1929... 100
1930... 93.14
1931... 79.27
1932... 83.71
1933... 83.19
1934... 88.00
1935... 87.96
1936... 90.80
1937... 93.32
1938... 91.08
1939... 87.54
1940... 85.95

The rise in the real price of gas was due more to general deflation than changes in the nominal price of gas. From 1929 to 1939 the CPI fell almost 20%, a much larger drop than the 13% drop in the nominal price of gas.

the nominal price of gas actually seemed to move with real gdp growth -- falling in 1931,32 and 1938 and surging in the years of strong real gdp growth --1934,35,36.

It looks to me the causal direction was much more from the real economy and general deflation causing swings in the real price of gas prices rather than in the direction you are suggesting -- from real gas to the economy

At 1/06/2009 12:21 PM, Blogger David said...

Hello Mark:

I get roughly the same trends you obtained by examining the "real" price of gasoline. This can be found at, suggesting that energy costs play critical roles in the overall health of the economy. Perhaps much more so than most people appreciate.


At 2/03/2009 1:39 PM, Anonymous Anonymous said...

Oil still at $24 1933? Over 3 years into the dowturn and ~90% reduction in stock market "a fairly ordinary recession." Use a different spin.

At 3/10/2009 9:08 AM, Anonymous Anonymous said...

That oil prices moved up with GDP growth (total demand) must have been a tough row to hoe in the 1930s; it suggests that there was little wiggle room in production capacity. Aren't we in a similar place in 2009? If economic growth (largely tied in with petroleum usage) occurs, then demand grows, but supply seems incapable of keeping up, so prices spike again, rinse, wash, repeat.


Post a Comment

<< Home