The Relative Performance of the Economy under the Presidents of the United States from 1900 to 2022

Revised October 15, 2023

Samuel H. Williamson

Founder of MeasuringWorth

 and 

Professor of Economics Emeritus Miami University 

sam@eh.net

Economic performance has been an important issue in elections for decades. Voters look at economic variables to gain some insight into the incumbent's record. The health of our economy, however, is determined by monetary and fiscal policy, external shocks, and even bubbles. The economy faced by an incumbent president and Congress, good or bad, is the result of events that occurred before as well as during his term of office. Elections have been won and lost by the timing of the business cycle where incumbents were caught on the right or wrong side of the slope. Of course, this does not stop claims and blames during election campaigns. Some are rhetorical, such as "the best economy ever" or "you are worse off than you were two or four years ago." Others are specific such as "your taxes are higher today than they were" or "the unemployment rate fell last month." In many cases, these statements have little context or are misleading. For example, taxes may be higher than four years ago, but at that time, there may have been a onetime rebate to deal with a recession. The unemployment rate may have fallen but from an unusually high rate. 

The goal of this paper is to help the reader better understand how the economy performed during each of the presidential terms from 1900 to 2022. It has been updated October 2023, so that data from the Trump term are complete and the first year of the Clinton, Bush, Obama, Trump and Biden administrations can be compared. Data on important aggregate economics measures are presented in the 11 tables and discussed in the text so that readers can be better informed as they make their own conclusions on these issues.

Sources 

MeasuringWorth is a website devoted to presenting the best historical time series of economic data and providing comparators that help make useful comparisons of these data. The tables presented here can be reproduced easily using data from the MeasuringWorth website, FRED, and other government sources. They can be studied using the site's Annualized Growth Comparator, which analyzes the growth of dozens of variables. The tables use annual data from 1900 to 2022 augmented by quarterly or monthly data from the post-World War II, the earliest the more frequent data were measured.

Comments on Data.

Gross Domestic Product (GDP) is the total market value of all final goods and services produced. The values posted are the measures of this flow in a quarter or a year.¹ This flow is not constant during the period. There are more goods and services are produced on a Tuesday afternoon than at 3am on a Sunday morning. A year can start in a boom, and being in depression by July. The annual observations will not reflect this change. Quarterly observations, do a better job of measuring changes, but they're still cannot date precise turning points.

 The Consumer Price Index (CPI) is a weighted measure of the cost of a bundle of things the average household buys. In theory it is the cost at a point in time and what we call a stock variable.² Tax rates, interest rates and the price of gold are all stock variables. While there are earlier contemporary measures of prices and employment, the accuracy is far less than it is today. 

 Another concern when making comparisons over time is the problem that these variables may have the same name, but they measure different things. GDP is a total product of each year and in 2022 millions dollars of smartphones were produced. In 1990 there were none. There was no agency measuring the GDP before WWII, so those observations have been estimated by historians. The typical household spends a large share of its budget on food in the 1930s. There was no internet and TVs in those days. 

 The data are reported by each president. A separate observation has been created for those terms in which a Vice President succeeded to the presidency. In four cases, this was a result of the President's death; in one case, the President resigned.³ For example, there is one observation encompassing FDR's first three terms, one for the term in which he died and Truman took over, and one for the term that Truman was elected. To make comparisons of terms of different length, we report most of the observations as annualized growth rates. For the annual data we have set the length of a Presidential term as change in levels from the year the President was elected, to the year of the next election. Thus we have four years of observations.⁴

 The quarterly data sets the term from the quarter before they were inaugurated to the quarter their successor is inaugurated. And for the monthly labor data, we've used the January observations of the year they took office to the January of the year they left office.⁵ To make comparisons of terms of different length we convert most of the observations to be the annualized growth rate.

 

Table 1

The first table presents annualized growth rates of key economic variables for the terms of 22 "presidents." It covers the inflation rate, the growth in the S&P stock index, population, the price of gold, real GDP, and real GDP per capita. Each variable has had its ups and downs (sometimes coincidental) with presidential terms.

Inflation, as measured by the rate of increase in the CPI, was highest during the Wilson administration because of World War One and the Roosevelt/Truman term because of the ending of World War II price controls .

The other period of high inflation is during the Nixon/Ford and Carter administrations when the Federal Reserve was increasing the money supply at a rapid rate. Ford and Carter were both criticized for not being able to bring inflation under control and both lost their elections of 1976 in 1980 consecutively. Biden is receiving the same kind of criticism, but the inflation rate in falling faster now that it did in those past years.

Stock prices rose during every administration, except Hoover's. Only during the Nixon/Ford and G.W. Bush administrations did the stock index grew slower than the inflation rate. We will have more on stock index comparisons in tables below.

Gold have only been a true market commodity for part of the time.⁶ It is easy to see the relationship between its price and the inflation rate during the Nixon Ford and Carter administrations. As inflation increased, gold was bought as a hedge against the falling value of the dollar. Gold prices and the stock market can be counter cyclical such as took place during the G.W. Bush administration.

Population generally grew fastest during the first years of the 20th century, up to the 1920s with higher birth rates and immigration. The rate of growth dropped during the Depression, then increased as a result of the post-World War II baby boom. During the 21st century there has been a continual decline and is now at its lowest rate ever with continuing falling birth rates, and restrictions on migration.

Economic growth, measured either as the growth in real GDP or real GDP per capita, is the most often used measure of a President’s success in managing the economy, whether or not they have had any control over the sources of growth.⁷ The next two tables present various rankings of GDP growth.

Table 2

This table ranks the Presidents by the amount of annual growth in both real GDP and GDP per capita. For each President there were particular events that influenced growth during their term. Some might have been their policies; some might have been the polices of their predecessor; and some might have been external events that were beyond their control.

Keeping these cautions in mind, we see at the top of the lists is FDR who took office near the trough of the Great Depression and died near the peak of a cycle near the end of World War II. Second to last is the Roosevelt/Truman.⁸ This can be explained partly by the drop of the government share of GDP from 41% to 14% between 1945 and 1947.

We also see that of the top seven administrations on the list, five are Democrats. While some of this is explained by business cycles, and war and peace, the higher ranking of Clinton and Carter administrations is not. Of the twelve at the bottom of the list, nine are Republicans.

Table 3

This table separates the growth in GDP into the share contributed by government expenditures, and that from the private sector. Note, these are government expenditures and do not include transfer payments such as the payments in the stimulus package.¹⁰

A perpetual platform of the Republican Party is to reduce the share of government in the economy. The Eisenhower and Nixon/Ford administrations were leaders in doing just that. Reagan and the Bushes were not able to follow this prescription as well and were worse at it than all Democrat administrations except Truman and Kennedy/Johnson. An increase in military expenditures shows up in the Truman administration with the Korean War, the Kennedy/Johnson administrations with the Vietnam War, the doubling of the defense budget during the Reagan administration,¹⁰ and the wars with Iraq by Bush presidents.¹²

One way to see how well the non-government economy is to compare its growth with the growth of the economy as a whole. If the non-government economy is growing faster compared to the total, the more the economy is growing without the lead of the government sector. The second part of this table shows the difference between the two. The larger the number, the more the private sector is leading the economy. We see that peace time presidents do the best. First it is Nixon/Ford, then Eisenhower followed by Clinton, Obama and Carter. Reagan, G.W. Bush and Trump are negative in this comparison as is Truman as was explained above.

Another thing revealed by this table is that for first two quarters of 2020, non-government GDP dropped by 12.6%¹³ while government share rose by 1%. This is a good counter cyclical measure whether it was intended that way or not.

Table 4

This table presents the interest and maximum tax rates by each administration. The interest rate used is the federal funds rate. The tax rates are the maximum individual and corporate rates.

Wilson and Carter suffered the largest increase in interest rates during their terms, and as we mentioned above, both faced high inflation rates. High inflation always generates higher interest rates. The rate increase in 1928 was the FED’s attempt to dampen stock speculation. The 2.72% increase in the Clinton term reflects the problem of using just the first and last observations. The 2000 interest rate was 6.24%, but the rate for 1999 was 4.97% and for 2001 it was 3.88%.

The story from the maximum income tax rates is consistent with political expectations; five of the eight times income tax rates were decreased, it was during a Republican administration. In only one case, Harding/Coolidge, does a Republican administration raise the maximum corporate tax rate. And in only one case out of the four times the maximum corporate tax was decreased was it done by a Democrat administration.

The largest increase in the maximum of both taxes rates, was to finance the two world wars, the largest decrease in the income tax was during the Reagan administration and the corporate tax during the Trump administration.

Tables 5 & 6 & 7

Table 5 shows the number of employees in the different sectors of the economy using monthly data. Table 6 gives the percentage increase in the number of employees in each category by administration and table 7 gives the ranking of that growth by the seven different measures,

The five top administrations in rank of growth in the total private employment are Democratic administrations. And with the exception of the Obama administration, the rest of the 14 ranked in this list are Republicans.

There were 18.4 million manufacturing workers at the end of the Carter administration of the 1980s, the largest total in the country’s history. This number declined for the next 50 years and was 12.3 million when Trump took office. Almost 80% of that decline took place from 2001 to 2009, the G.W. Bush term when globalization was in vogue. During the Trump administration there was an increase of of about a quarter of a million jobs in manufacturing, through all happened during the recovery from the COVID epidemic.

Government Employment is a bit paradoxical. The maximum number of Federal Government employees was 3.1 million at the end of the Reagan administration and currently as of August 2020. These are both Republican administrations. The largest decrease in Federal Government employment was a quarter million workers during the Clinton Administration.

State and Local Government employment saw steady increases from the time of Truman to the end of the 20th century. For this century, the employment of these governments is down about one percent while to population has increased fifteen percent.

The unemployment rate is not ranked in this essay because the observations have considerable “noise” in them; the official definition of unemployment has changed many times over. Further, the participation rate, the number of discouraged workers, layoffs or on those on strike all can impact this rate. Employment levels are better measures of the economy’s health.

Table 8

This table presents the total federal debt at the end of each president’s term in $ billions from 1900, and the debt held by the public from 1944 to the present.¹⁴ The last column presents its annualized growth during their years in office. The debt goes up when the government increases its spending or has a reduction in revenue. The latter happens in recessions when tax revenue falls with the fall in incomes, or when tax rates have been decreased. Spending is determined by Congress in its annual budget process.

It is not surprising that the largest increases were during the terms of Wilson and FDR who were financing the two World Wars. Between the wars, the attitude was to reduce the debt by running surpluses to pay off the war borrowing. After WWII, the same attitude prevailed and in January 1956, Eisenhower was to say "In this connection, I should mention our enormous national debt. We must begin to make some payments on it if we are to avoid passing on to our children an impossible burden of debt."¹⁵ For the next two administrations the growth was back to the peacetime levels. After that the growth rate went up, however, when Carter left office it was at the low rate of 32% of GDP, the lowest level since the Coolidge administration.¹⁶ It is now 156%.

Eisenhower was opposed to any tax cuts until the budget was balance. Kennedy proposed a tax cut that Johnson was able to get passed, and the economy was growing so fast (as we saw in Table 1) that “it paid for itself.”¹⁷ Ever since tax cuts have been promised that they will pay for themselves and it has not happened. Reagan, G. W. Bush and Trump are known for important tax cuts and corresponding increases in the growth of the debt.

The largest post-war deficit increase was during the Reagan administration.¹⁸ The exception for the rest of the years of our study was during the Clinton administration where there were budget surpluses and the annualized growth rate of the share of the debt held by the public dropped from 21.5% to less than 1%. Eisenhower would have been pleased with what Clinton accomplished. For the entire twenty-first century the U.S. Federal debt has been growing at over an average of 8% a year.

Tables 9 & 10

It is often stated, “the stock market is not the economy.” In general the thought is that the value of an individual company’s shares reflects its expected earnings in the future and the price goes up and down with those expectations. . The value of stocks is influenced by such things as changes in investors’ expectations of changes in interest rates, consumer confidence, planned and expected government outlays, and tax rates.

If a new President is regarded as good for the stock market, we would expect that this will be reflected by faster increases in the indexes than a President who is regarded as a danger to the market. This impact can start the day they are elected or the day they are inaugurated. Many people make comparisons each way, so both are presented here.

During their terms all presidents have had a day where the market was higher than the day they were elected and inaugurated.¹⁹ Ten months after Herbert Hoover was elected the DJIA had increased 46%. Investors were pleased with their new president. On the last day Hoover was in office, with the economic deep in depression, the DJIA stood at 20% of its value compared to four years earlier on his election day.

How does this compare with other Presidents? Table 10 shows the highest value of the DJIA during each of the 30 terms of all the Presidents from 1900 to 2020 and compares it to the value on the day they were elected and the day they were inaugurated. We do not consider when in the term it takes place with some peaks coming early in term and others on the last day. We are just seeing that at some point in his term the market had peaked and how much was that.

The rankings are interesting in that FDR has the largest increase in his first term of over 300% since he was elected and inaugurated. His second term shows basically no increase. Bill Clinton was the standout being third and fourth out of 30 in having a higher value since he was elected. He is third and sixth since he was inaugurated. Reagan performance was mediocre in his first term, but for his second term he was second in both lists to FDR with a higher value of over 200%.

Conclusion

As was said at the start of this essay, the performance of the economy during a president’s term, good or bad, is the result of events that occurred both before and during his term of office and that were often beyond his control. This can make it difficult to determine how much to credit or blame the president for what occurred.  But it is done all the time, particularly during election times, so we hope these data provide a more accurate way to make such evaluations.  It is likely that most economists might think that it would be more insightful to use economic performance to rate the chairs of the Federal Reserve Board.

¹ To measure GDP takes hundreds of national income accountants working year round. There are three versions presented during the year.
Back to text

² Tax rates, interest rates, stock indexes, population and the price of gold are all stock variables.
Back to text

³ We do not separate the first term of Theodore Roosevelt when he took over after the death of William McKinley.
Back to text

⁴ We could have used the year after he was elected to the year his successor takes over. The theory in doing it that way is that it takes a year for a president’s policies to have an impact. Table 3a versus 3b does this comparison and we did construct many of the tables with annual data that way and while it did change some rankings, it was not significant. If you want a copy of those tables, please contact the author.
Back to text

⁵ From 1952 to the present that is the month of the inauguration. Before that, the date was March 3rd.
Back to text

⁶ From 1932 to 1968, gold was held by central banks and Americans were not allowed to own it.
Back to text

⁷ These total output do not include transfer payments such as the payments in the stimulus package.
Back to text

⁸ Truman took over (less than three months after Roosevelt died into his fourth term.)
Back to text

⁹ Obama came into office at the peak of the Great Recession; output continued to fall until the end of 2009. The difference in his ranking between 3a and 3b shows this.
Back to text

¹⁰ Transfer payments such as Social Security, Medicare and unemployment benefits are not payments for production and are not part of GDP.
Back to text

¹¹ The Trump administration has had a measurable increase in the defense budget as well.
Back to text

¹² In this table we do not separate the two administrations where a vice president from the same party took over.
Back to text

¹³ 12.6% = (15,917-13,911)/15,917.
Back to text

¹⁴ This is the debt excluding what is held by other government agencies such as the Social Security Trust fund and the amount held by the Federal Reserve System.
Back to text

¹⁵ On that day, the per capita debt was $1,600, today it is over $200,000.
Back to text

¹⁶ Remember that these data are in nominal terms, so the inflation of this period has an impact on this ratio.
Back to text

¹⁷ A lower tax rate times a larger income can be generate more or less revenue. For the Kennedy/Johnson administration, revenues went up. It was a relatively small cut and the Federal expenditures we also cut.
Back to text

¹⁸ There was a recession and a large tax cut during the eight years of his administration that caused the rapid increase in the debt.
Back to text

¹⁹ That was less than 10% in eight cases of the 60 we examine.
Back to text

---

Please let us know if and how this discussion has assisted you in using our calculators.