Glenn Beck spent a portion of his show on February 9, 2010 discussing President Coolidge and the Roaring Twenties. I analyzed his claims about Coolidge in an earlier blog post, and have returned to fact check his points on the Roaring Twenties.
He begins by calling the Roaring Twenties “arguably the most prosperous 8 years this country has ever seen,” credits Secretary of the Treasury Andrew Mellon’s tax cuts, and cites innovation as evidence of this prosperity.
In 1920, there were only 5,800 people who had ever flown on an airplane. By 1930, it was seventy times that amount. RCA changed the world with the radio. Along with the radio came another invention or another idea: advertising. Now you could hear Babe Ruth hitting home runs anywhere in the country while someone was telling you about a product. Thomas Edison brought us movies in 1880s. You know when they really started to take off? In the 20s the true modern era motion picture arrived.
And so on. But were the Roaring Twenties really all Beck makes them out to be?
First, let’s deal with length. Beck notes that the economic expansion of the 1920s lasted 8 years — how does that compare with other U.S. expansionary periods? Of course, this depends on who you ask. For instance, the Post World War II economy could last from 1945 to 1960 or until the early 1970s. The Reagan boom beginning in 1982 could last until 1990, or it could have been part of a much broader boom that included the Reagan, Bush, and Clinton economies, ending in 2001. James Pethokoukis, writing in the U.S. News and World Report, even goes so far as to proclaim that this expansion lasted from 1982 to 2007 — that’s 25 years!
Of course, these booms were not uninterrupted. There were recessions in the 40s, 50s, and 60s, and again in the 80s, 90s, and 00s. Perhaps the 1920s were the United States’ longest uninterrupted boom? Not quite — there were recessions during the Roaring Twenties as well.
A peek at the National Bureau of Economic Research’s website, which has an excellent list of the U.S. economy’s expansions and contractions, shows that the U.S.’s longest uninterrupted expansion occurred from March 1991 to March 2001, from the end of the first Bush Administration through the Clinton years and into the beginning of the second Bush Administration. The second longest uninterrupted expansion occurred from February 1961 to December 1969 (the Kennedy, Johnson, Nixon years), the third from November 1982 to July 1990 (the Reagan and Bush years), and so on. The longest uninterrupted expansion during the Roaring Twenties period lasted from July 1924 to October 1926 — not even in the top 10 for length of uninterrupted booms.
Clearly, the Roaring Twenties were not the longest boom in American history, but (as with all aspects of life) quantity and quality do not always correspond. So how about a comparison of content?
Let’s compare some indicators with similar ones from the three longest booms in American history (1991-2001, 1961-1969, and 1982-1990). First, we’ll look at the obvious — Gross Domestic Product (GDP) (the country’s economic output).
And the unemployment rates…
Thus, GDP growth was better in the three booms we’ve been examining than in the Roaring Twenties, while unemployment seems to have been lower in the Roaring Twenties. Ironically, the minimum wage may be to blame for at least part of the higher unemployment rates of the later three booms. The minimum wage was first established as part of the National Industrial Recovery Act of 1933, which was declared unconstitutional by the Supreme Court. It was then re-established in 1938.
One of the economic consequences of the minimum wage is to increase unemployment. Think of it this way: a business owner has a certain amount of money available to spend on employees. If the amount he must pay his employees rises, but the amount of money he has available to pay them does not, he must fire someone. This force was absent in the 1920s, perhaps partially accounting for the lower unemployment rate.
Another aspect of the booms is income inequality. One of the ways this is measured is known as the GINI Index. The Gini Coefficient measures income inequality on a scale of 0 to 1, with 0 being absolute equality and 1 being absolute inequality. The following graph displays Gini estimates for the 1920s to the 1990s.
Unfortunately, the late 1990s and 2000s are not included in this graph, but the data shows that the GINI Index for families continued to rise, and even reached its highest reported point yet in 2006 at 0.444. As the graph shows, GINI estimates for the 1920s start somewhere around 0.425 and then quickly rise throughout the decade, peaking at around 0.5 before relaxing to 0.45 and then rising again to about 0.475 to finish.
This high level of inequality stands in stark contrast to the much lower levels enjoyed during the 1961-1969 boom, where the GINI Coefficient began at 0.374 and finished at 0.349. Indeed, the lowest reported level of income inequality ever occurred in 1968, at 0.348.
The 1982-1990 boom featured greater inequality than the 60s, but still lower than either the 20s. The GINI Coefficient began at 0.380 in 1982 and stayed in the high 0.3s for the duration. It peaked at 0.401 in 1989 and finished at 0.396 in 1990.
Family income inequality increased yet again during the 1991-2001 expansion, starting at 0.397 and climbing to 0.435 at the end, its highest year. This 2001 level is higher than estimates from the beginning of the 20s, but lower than most other years of the Roaring Twenties.
Emmanuel Saez examined the share of total income of top decile (that is, the top 10th) and also the top 0.01% of earners in his paper, “Striking It Richer: The Evolution of Top Incomes in the United States.” As the graphs show below, he found that the top 10% of earners held almost 50% of total income in the United States by the end of the Roaring Twenties. This is almost matched by the high level of income held by the top 10% at the end of the 1991-2001 boom (and exceeded in 2007, as an aside). The 1960s boom seems to be the most equitable, followed by the 1982-1990 boom, which saw increasing inequality, but not at the levels of either the Roaring Twenties or the 1991-2001 boom. Examinations of the share of the top 0.01% yield similar conclusions.
One of the points Beck makes when discussing the prosperity of the Roaring Twenties is the innovation associated with the period. He appears to describe innovation in terms of something once great, now long lost. Yet innovation remains a strong part of our capitalist system. Beck awes his audience by telling them about the great leaps in technology that were made in the 1920s, and how the long-lost capitalism of Coolidge and Mellon made expensive technologies affordable.
But aren’t these processes still occurring today? Consider, for a moment, computers. In 1984, 8.2 percent of U.S. households had a computer. By 2000, 51 percent of households had computers. That’s an increase of a whopping 522 percent! This is how capitalism works — competition drives down prices. The Roaring Twenties do not hold a monopoly on innovation and advancements in technology.
Of course, other than simply listing the accomplishments of these various years (which are many) and the price drops as technology advances, how does one measure innovation? I’ve sorted through copyright registration and population figures to derive some glimpse at innovation in these various booms. The following graph lists the number of copyright registrations in the beginning and ending years of each boom, the populations of those years, and then the registration figures divided by the population (revealing how many copyrights were registered per person in the United States for those years).
Not only do the number of copyright registrations rise with each successive boom, but they also rise per person — a real increase. This is with the notable exception of the 1991-2001 boom, in which copyright registration figures actually decrease in per person terms, though they remain above figures for the Roaring Twenties and the 1961-1969 boom.
Productivity gains, too, may be some indicator of innovation, as new technologies make manufacturing more efficient.
These estimates show productivity gains in each of the four booms, with the average yearly percentage change in productivity in the Roaring Twenties and the 1990s booms being almost a full percentage point higher than those of the 60s and 80s booms.
Perhaps we can see, through these somewhat rudimentary lenses, the steady progression of innovation.
As far as comparing booms goes, a paper by economist Robert J. Gordon notes the similarities of the 1990s and 1920s booms:
Growth in real GDP, real GDP per capita, employment, and productivity were almost identical, the conventionally measured unemployment rate was identical in 1928 and 1999, inflation was negligible (1920s) or low (1990s), and the late-1920s stock market boom is the only such episode in the century that comes close to the stock market’s ebullience in the late 1990s. Like the 1990s, the 1920s witnessed prosperity, a productivity revival, low unemployment, and low inflation. Both decades featured an explosion of applications of a fundamental “General Purpose Technology,” electricity and the internal combustion engine in the 1920s and computer hardware, software, and networking communications technology in the 1990s. Both decades appear to mock the existence of a Phillips-curve tradeoff between inflation and unemployment.
Of course, these figures are limited in that they are solely economic indicators. Bobby Kennedy noted the inadequacy of such measurements in 1968:
Yet the gross national product does not allow for the health of our children, the quality of their education or the joy of their play. It does not include the beauty of our poetry or the strength of our marriages, the intelligence of our public debate or the integrity of our public officials. It measures neither our wit nor our courage, neither our wisdom nor our learning, neither our compassion nor our devotion to our country, it measures everything in short, except that which makes life worthwhile.
To provide a broader picture of the four booms, I’ve examined three other indicators — life expectancy gains, 5 to 7 year old enrollment (in school) as a percentage of the 5 to 7 year old population, and high school graduation rates and gains.
As would be expected, steady gains were made over time in life expectancy. However, averages of yearly percentage change show that the greatest gains were made during the 1991-2001 boom (0.22%), and the worst gains were made during the 1961-1969 boom (0.05%). The averages show a decrease actually occurred during the Roaring Twenties (-0.52).
As seen from the graph, the lower levels of enrollment in the Roaring Twenties (about ten percentage points below the 1980s/1990s levels of enrollment) also provided greater room for improvement, and, indeed, the Roaring Twenties saw greater gains in 5 to 7 year old enrollment as a percentage of that population (1.27%, on average). The 1960s boom also saw a high level of improvement, on average, as did the 1980s (the data for the 1980s also shows the movements of the baby boom generation out of secondary school and into college). Finally, the 1991-2001 boom, although maintaining a high level of enrollment in the 97%-98% range, saw a decrease in the yearly percent change of enrollment, on average (-0.08%).
(Quick note — the rounding to one decimal place should account for the oddity that occurs in 1967-1968, when the percentage change is -0.05 while the graduates as a percentage of the 17 year old population appears to stay the same).
Again, the low levels of the Roaring Twenties (about 50 percentage points below 60s levels) provided a greater room for improvement (6.55% yearly percentage increase on average) — and it took advantage of that room, increasing the percentage of the 17 year old population that graduated from high school every year of the boom! Although the other booms saw higher levels of graduation, none of them can boast continuous improvement (the 1960s boom saw great gains, as well, but also some declines). In fact, the 1991-2001 boom saw a steady decline, before leveling out just below its beginning 1991 level.
So the picture, on a whole, is mixed. Perhaps the comparison itself is fundamentally flawed because of vast differences of the times. The Roaring Twenties were not the longest boom in American history, nor were they greatest, in terms of GDP and output. They saw the highest levels of income inequality out of the four booms we’ve examined (though the 1991-2001 boom comes close) but also a lower level of unemployment. The life expectancy retreated a bit, on average, though schooling enrollment and graduation made real gains. Productivity in the manufacturing sector, also, was high, though lower than the 1991-2001 boom, and copyright registrations were lower per person than in the three other booms.
The true picture is complicated, and for Beck to frame the Roaring Twenties as the high water mark of United States economic history (“arguably the most prosperous 8 years this country has ever seen”) is simplistic and dishonest. Nevertheless, the Roaring Twenties were a prosperous period in United States history that, according to the data examined, saw numerous gains in various aspects of life.