data economics

The American Presidents….By the Numbers

In 1948, historian Arthur M. Schlesinger conducted a poll of 75 historians asking them to answer the question, who are our greatest American presidents?  Since then, and probably before, it’s been a pretty popular debate.

Though previous attempts to rank our presidents have been based on observation and opinion, the names at the top and the bottom of the lists are pretty consistent; Lincoln, Washington, FDR at the top. Andrew Johnson, U.S. Grant, Warren G. Harding are at the bottom. When you actually think about it though, it’s really an impossible question. Is Washington better than Lincoln? Probably not at winning the Civil War at least.  He owned slaves and had a penchant for leading uprisings against his government. Washington probably isn’t your guy in 1861. We’ll never know what speech Millard Filmore would have given the day after the Japanese attacked Pearl Harbor. Most of us will never know anything Millard Filmore said. I promise you though, he was our 13th president.

Circumstance plays a big part of it and so it becomes an illogical discussion.  So why do we do it?  Partially because debate is an inescapable function of the human condition. It’s what we do. We argue about just about anything we can compare. There is another reason though. A more practical one. We debate who our great American presidents are because we are bench-marking our current prospects. We are looking at the greats of the past to see the potential in our future and so we find value in the debate. Because we find value in the debate doesn’t mean that we are having the right debate though.

A more effective question is what makes a great president? And if we were to look at how we’ve evaluated them in the past the answer is clear. For the most part, it’s crisis.  The true great ones, Lincoln, FDR, Washington, led through periods of dire crisis.  Which begs the next question.  Do we really want the next great president? It probably sounds counter-intuitive but I hope I never live to see our next great American president.  I hope that great American presidents are over.  Great American presidents mean war, suffering, economic catastrophe and death.   So when we talk about great president’s, might we have the debate differently? Shouldn’t we shy away from desiring a great president and instead focusing on desiring a great presidency?  That’s what I’m rooting for.

With that in mind, we took a shot at looking back at our presidents through that lens.  We built an algorithm to see what the data says and then compared it to what history says.  The results were interesting.

The good news is that this is actually pretty easy.  Because you can  measure prosperity, progress and opportunity, things we would likely agree make for a “great presidency” by observing patterns in economic data, scope and scale of war, territorial expansion and Constitutional legislation.   So that’s what we did. The table below illustrates the rankings of American presidents by using an algorithm incorporating historically collected data.  The table compares the outcomes of the algorithm to the aggregated results of a dozen or so polls of historians regarding presidential rankings.   As you might expect, the data tells a different tale then the anecdote.

Score Algorithm Ranking Historical Ranking
100 George Washington 1 3
74 Ulysses S. Grant 2 37
73 Thomas Jefferson 3 4
69 Andrew Jackson 4 8
67 James Monroe 5 14
65 Bill Clinton 6 21
63 Ronald Reagan 7 17
63 James Madison 8 13
62 Franklin D. Roosevelt 9 2
61 Grover Cleveland 10 19
60 Rutheford B. Hayes 11 25
58 Dwight D. Eisenhower 12 9
57 Theodore Roosevelt 13 5
57 John Adams 14 18
57 William McKinley 15 20
52 Harry Truman 16 7
52 Andrew Johnson 17 41
52 James K. Polk 18 10
51 John Tyler 19 36
51 Barrack Obama 20 16
50 Lyndon B. Johnson 21 15
50 Woodrow Wilson 22 6
49 John Quincy Adams 23 18
48 Millard Fillmore 24 39
47 James E. Carter 25 27
47 George W. Bush 26 33
47 Calvin Coolidge 27 31
46 William Taft 28 22
45 George H.W. Bush 29 23
43 Franklin Pierce 30 40
42 Martin Van Buren 31 24
40 Chester A. Arthur 32 28
40 Richard M. Nixon 33 32
39 James Buchanan 34 42
37 Warren G. Harding 35 43
37 Benjamin Harrison 36 38
36 Gerald R. Ford 37 26
35 John F. Kennedy 38 11
22 Herbert Hoover 39 30
22 Abraham Lincoln 40 1
19 Zachary Taylor 41 35
17 William H. Harrison 42 38
14 James. A. Garfield 43 29

The data behind the comparison shows several things.  First, the algorithm and the historian polling are moderately correlated, meaning that the two lists are not entirely at odds with each other.  Immediately, some clear differences jump out at us though.  Here are some of the more glaring insights.

Where’s Lincoln? 

I challenge you to find a historian that does not include Lincoln in their top three on their index of presidential greatness. This algorithm, however, does not measure personal greatness.  It measures outcomes relative to the quality of life of the people being governed. Lincoln’s presidency was marked by unprecedented carnage through war, national crisis and ultimately assassination. It’s safe to say if we could have avoided it, we would have.  The numbers clearly show that, giving him the most significant historical overvaluation relative to the data.

There’s something else interesting though. You can’t really capture, through data, the accomplishment of paying off the debt of overdue societal progress. Which tells us that ignoring required change, like abolishing slavery, ultimately results in really lousy outcomes for the people who actually put their foot down to change it.  And though history treats them well, the lives of the American people, as they lived them, were miserable.  So if you can, change things before you have to.

Was Washington really that great?

Was Washington really worthy of the title father of our country?  The data says so.  He had the highest average economic growth of any two term president outside of FDR.  Despite our fledgling status as a nation and our relative inability to defend ourselves against foreign enemies, Washington managed to steer us clear of war.   He signed just under half of all Constitutional Amendments ever passed and he expanded the territorial holdings of our country from nothing to something.  The first eight years of our country’s existence could have gone terribly wrong but it did quite the opposite.  Washington oversaw prosperity, stability, growth and progress on a scale not duplicated by any president since.

Did we really get Grant that wrong?

Grant was a great general, but a bad president.  That is what I was taught in history class growing up and obviously what our historians voted as they ranked him the 37th ranked president out of 43.  The data shows something different.

Though recession hit during the latter years of his two terms, the recovery and post-Civil War boom actually show that America experienced 5% GDP growth annually during Grant’s two terms. This ranks him fourth among all two term presidents behind FDR, Washington, and Jackson for economic growth.  As a modern frame of reference, Reagan and Clinton, both uniformly considered to be fiscal successes as presidents, were both about 4.1%.

President Grant also governed during a period of relatively stable peace and even ratified the 15th Amendment providing the right to vote to African Americans, a significant political debate of the time. So why is history down on Grant? The headlines point towards corruption and the eventual recession of the mid 1870s. Data doesn’t measure corruptions.  Just outcomes, but it does raise an interesting question.  Should we care about corruption if it doesn’t hurt us?  I think we do but perhaps the lens is that it is more of a long term problem.  We shouldn’t tolerate corruption, even in an environment of prosperity because it erodes the fabric of our political discourse. And ultimately it breaks down.  Site Bill Clinton and the damage his character issues did to the perception of trust in our politicians.  More on him later though.

There’s something else interesting in the data relative to Grant. If you look a little deeper, we begin to see indications of what may have been influencing our historians in their selections. Of the presidents that have the largest historic undervaluation relative to the algorithm, the top two, Grant and Andrew Johnson immediately followed Lincoln.  Rutherford B. Hayes, who followed Grant, also cracked the top 6 of undervalued presidents. This group who ushered in the era of those labeled the “forgetful presidents” has been much maligned by history.  But they actually governed during a period of unprecedented growth and stability.  But we know growth and stability isn’t what we remember. We remember the other stuff.  It’s safe to say that Grant, along with Andrew Johnson and Hayes suffered mainly from a case of not being Abraham Lincoln.  History has never really gotten over the fact that America was robbed of being led through post-Civil War reconstruction by the hero that delivered us from near destruction as a nation.

What gets a president noticed?

James Garfield spent 200 days in office.  Despite that, historians have him ranked as the 29th greatest president.  That means, despite being in office for about as long as a single Major League Baseball season, Garfield is considered a more effective president than 14 other men by historical opinion.  Only two of those 14 also served for under a year. Which means Garfield, having done nothing at all, in a literal sense, is considered more effective than 12 other presidents who served in office for years. This includes two term Presidents Grant and George W. Bush.

How is that possible?  The pattern in the data is very clear.  Of the four most overvalued historic presidents, three were assassinated.  Lincoln, Kennedy and Garfield were all killed while in office.The fourth most overvalued president, Woodrow Wilson, was a war time president.   If you want the American people to remember you fondly, get killed in office or go to war. Both things, most would agree would be outcomes to avoid, if we could.


What about the new guys?

The algorithm doesn’t care about how recently you were president. Reagan and Clinton are both ranked in the top ten, having served two peaceful terms of economic prosperity within the last 40 years. Both also crack the top eight undervalued presidencies. Historians tend to need some water under the bridge before they feel justified in giving due credit. After all, they are historians.  The data says that President Obama is ranked 20, just four rungs below where our historians forecast his placement.  The data behind his predecessor, George W. Bush, hands him the title for the lowest ranking full two term president at 26. I have some particularly leftward leaning friends who wonder regularly how “W” got two terms. The data supports their concerns. To keep this discussion data driven and bi-partisan, the outcome of the “Hope and Change” promised by Candidate Obama has him looking up at President John Tyler. From a data perspective, Tyler had a more effective presidency despite his somewhat less inspiring campaign slogan of “Tipecanoe and Tyler too…”

So what?

In the end, the data is just another way to debate the question.  It’s an algorithm that was built by a person, which means it is subject to its own biases and inadequacies. What it does show is that there are patterns to our bias that data and analysis can point out. It also shows that data, while important, often misses the qualitative aspects of measurement, like the gross injustice of slavery that mandated Lincoln’s great national and personal sacrifice.  Or the scandal and clear instances of dishonesty of the Clinton era and the long term erosion of the confidence in government.  But the data does serve to offer a different perspective. It’s why we use data in business.  It’s why we use it in sports.  And now, more than ever before, data is how we make sense of our past and the world around us.

For me it brings our two critical insights.  And the first is that presidential performance is outlived by societal impact.  You can change, for good or bad, things that long outlive your term.  The second is that change is easiest before its needed.  And though we celebrate the presidents who force it under dire circumstance, the lives of Americans who lived through it are largely miserable.  So change things before you absolutely must.  Think social security, climate change etc.  If you don’t and you rely on the “great man” to do it for you, the fee is high for that service.

In the end, it’s data.  And data helps start the discussion.  If it ends there, it’s less useful.  But if you ignore it, you tend to start in the wrong spaces.