This professor has predicted every presidential election since 1984. He's still trying to figure out 2016.
Allan Lichtman says he can predict the outcome of any US presidential election. He often does it months or even years ahead of time. Oh, and his predictions have been right in every presidential election since 1984.
But Lichtman, a distinguished professor of history at American University, doesn't use polling, demographics or sophisticated analysis of swing states. He makes his predictions based on 13 true/false statements that he says indicate whether the incumbent party will retain the White House or lose it in a given election.
Lichtman and Russian scientist Volodia Keilis-Borok came up with the keys - a series of true/false statements - in the early 1980s. The idea is that if more than half of the keys are true, the incumbent party will stay in power, and if more than half are false, the challenging party will win the White House.
Professor Allan Lichtman has correctly predicted the outcome of every U.S. presidential election since 1984. Here's how. (Peter Stevenson/The Washington Post)
The keys, which are explained in depth in Lichtman's book "Predicting the Next President: The Keys to the White House 2016" are:
1. Party Mandate: After the midterm elections, the incumbent party holds more seats in the U.S. House of Representatives than after the previous midterm elections.
2. Contest: There is no serious contest for the incumbent party nomination.
3. Incumbency: The incumbent party candidate is the sitting president.
4. Third party: There is no significant third party or independent campaign.
5. Short-term economy: The economy is not in recession during the election campaign.
6. Long-term economy: Real per-capita economic growth during the term equals or exceeds mean growth during the previous two terms.
7. Policy change: The incumbent administration effects major changes in national policy.
8. Social unrest: There is no sustained social unrest during the term.
9. Scandal: The incumbent administration is untainted by major scandal.
10. Foreign/military failure: The incumbent administration suffers no major failure in foreign or military affairs.
11. Foreign/military success: The incumbent administration achieves a major success in foreign or military affairs.
12. Incumbent charisma: The incumbent party candidate is charismatic or a national hero.
13. Challenger charisma: The challenging party candidate is not charismatic or a national hero.
So how does all of this apply to Donald Trump and the wholly unusual 2016 election? Lichtman is still trying to determine his prediction.
*Clinton goes after Donald Trump on taxes, character
* Poll finds Trump, Clinton even as they turn to election
* Sanders, Trump come out on top in West Virginia
* Clinton, Sanders to face off in West Virginia
* Trump has one strategy for defeating Hillary Clinton
* Congratulate me, Trump demands of CNN journalist
We sat down with Lichtman at his Washington office this week to get his thoughts on the 2016 race and how it might play out. Our conversation has been edited only for length.
Q: For folks who may not be familiar with the book or the keys, can you tell me how you began to approach this and where the idea that you might be able to predict presidential elections based on factors, like how the incumbent president is doing, come from? How did that whole idea coalesce into a measurement system that you can use to predict an election?
LICHTMAN: Well, we first developed the keys in 1981, and they have since predicted correctly the popular vote in all eight American presidential elections, from 1984 through 2012 - usually years ahead of time. I predicted the very-hard-to-call 2012 election in print in 2010.
When you think about it, predicting elections is much like predicting earthquakes. You're predicting whether there is going to be stability - the party holding the White House keeps the White House - or an earthquake - the party holding the White House loses, and the challenging party wins.
So we reconceptualised presidential elections, not as Carter vs. Reagan, Republicans vs. Democrats, liberals vs. conservatives, but as "party holding the White House" vs. the challenging party. Now, in earthquake prediction, they predict by looking at factors in the physical environment associated with stability and upheaval. So we chose to look at factors in the political environment associated with stability - the party holding the White House stays in power - and earthquakes - the party holding the White House is thrown out.
And I had a theory behind this. This did come from my studying of presidential elections. And my theory was that the pundits and the scholars are all wrong about predicting presidential elections. That the real key is not the candidates, or the issues, or the debates, or the ads, but rather the performance of the party holding the White House - that essentially, American voters are ultimately pragmatic. And if the party holding the White House did a good job, they'd give them four more years. If not, they'd toss them out.
Now, that's just a story, not yet a model. But we created a model using that theory, by looking at the political environment in every American presidential election from 1860 - the horse and buggy days, when Abraham Lincoln ran against Stephen Douglas - to 1980, the modern era of television, polls and jet planes, when Ronald Reagan ran against Jimmy Carter. And we came up with a model which involved 13 keys - 13 key factors. These are simply true-or-false statements that can be answered prior to an upcoming election. They test the political environment, and they're primarily based upon the performance and the strength of the party holding the White House.
And we came up with a decision rule, a very simple one: If six or more of the 13 keys went against the party in power, that is, the answers to the questions were false, the party in power lost. If fewer than six keys were false, the party in power won. And that held, retrospectively, for every election over 120 years.
Q: When you talk about the strength - the relative strength - of the incumbent party, does that to an extent reflect the mood of the country and the public perception of that incumbent?
LICHTMAN: I don't use any polls. The conventional predictors use polls and other factors like the economy. The problem with conventional predictions, based on conventional statistical methodology, is that you've got to multiply presidential approval, say, or growth in the economy, by some multiplier. Now, that multiplier works great for past elections, when you know what the answers are, but the multiplier unpredictably changes for a future election. We don't do that at all. We have 13 factors; they all count equally. We have no multipliers that have to change from election to election.
Now, I've heard people say to me: "You're weighting these equally? How can you weight equally the economy during the Great Depression, which only counts for two keys?"
And this is the secret to the system. It's called trigger effects. Without having to weight any of the keys, if one or two keys are important enough, they'll trigger other keys. So the collapse of the economy triggered losses for Herbert Hoover's Republican Party in the midterm election of 1930. It triggered social unrest. It triggered a charismatic Democratic candidate who we all know, Franklin Delano Roosevelt, to run in 1932, when he had never been planning to. So three more keys are triggered, so we don't have to weight the keys and make mistakes.
Q: What are your current predictions for 2016?
LICHTMAN: I don't have a prediction yet. It's looking shaky for the party in power, but the prediction is not yet set because there are still two uncertain keys, and there is also a third possibility, which is not strictly a key but I talk about it in my book, and that is the challenging party dividing itself. So even if the keys might favour the challenging party to get a majority of the popular vote - if they split it among two candidates, they're still going to lose. That's very rare. Usually you get a third party who's truly independent, like Ross Perot, or a split within the party in power, like you got in 1912 when Teddy Roosevelt ran as an independent, former Republican president against Republican president William Howard Taft.
So we could see something quite unprecedented here. We just don't know yet.
Q: We often know who the candidates are going to be well in advance of the conventions, and you talk about how the party being split is this kind of unknown factor. How does that change things? Does that have more weight in a year like this?
LICHTMAN: It does, and it would specifically change things if the Republicans found a third party or independent candidate to run. We don't know if that's going to happen. There's a lot of talk about that, but that would be an extraordinarily unique situation of the challenging party dividing itself and running two separate candidates.
Q: One of the keys is "scandal," and this has been a particularly nasty election in which both current front-runners have particularly high unfavorable ratings, like Clinton's email server or Trump University. What threshold does that have to reach to really become a very influential factor?
LICHTMAN: So when we're talking about scandal, we're talking about presidential scandals. We're talking about the Watergate scandal, the Clinton impeachment, the Teapot Dome bribery scandal of the 1920s. And there is nothing remotely close to that for the Obama administration. Candidates trying to undermine each other, that is as old as the Republic.
These personal charges and countercharges are really the common coin of American politics. They ebb and flow; they're more severe at some times than others. They're pretty severe now, but they've been more severe in the past.
Q: Key 13 is "challenger charisma." The man challenging for the non-incumbent party seems to have a level of charisma we've rarely seen in American politics, even when you talk about Bill Clinton in 1992 or Obama in 2008. Clinton played a little saxophone. Donald Trump was a celebrity in his own right. Can that challenger charisma help fire up a base in a way that really swings things in their favour?
LICHTMAN: The charisma key has two parts. One is, you've got to be a national hero, like Dwight Eisenhower or Ulysses Grant. Obviously Donald Trump does not fall into the category of national hero. You know, running a reality TV show is not the same as winning the Civil War or World War II.
Now, the problem is, unlike candidates that I have previously characterised as charismatic - Teddy Roosevelt, Ronald Reagan, John Kennedy - his charisma is limited to a very small base. He has a small base of dedicated followers, but he also has the biggest negatives of any presidential candidate in the history of modern polling. So you've got to have charisma that extends beyond your narrow base to win that key. Donald Trump does not win that key.
Q: Is there anything about this election in particular that makes evaluating your keys and coming to a final prediction more difficult than it's been in other years?
LICHTMAN: This is perhaps the most difficult prediction I've had since 1992, because the prediction in 1992 depended on the party in power losing the "third party" key to Ross Perot, who was in the election, then out of the election, then back in the election. It was very hard to call.
This election is extremely hard to call for a number of reasons. No. 1, the contest key. Hillary Clinton has been the presumptive nominee, quite frankly, for some time. There was no way Bernie Sanders could catch her. But he's persisted. He hasn't given up. And the question is, ultimately: Will she win enough delegates to say the Democratic Party wins the contest key? You can't say that yet.
Another key that's very much in the balance is the "foreign policy success" key. Now, most experts in the field believe that the Iran nuclear treaty is a huge historical success, at a minimum averting a nuclear arms race in the Middle East, and at a maximum averting a nuclear war in the Middle East. Remember, this treaty is not designed to make Iran a good actor, it's not designed to stop their support of terrorism, just as the treaties with the Soviet Union were not designed to make them, you know, a better country, but to achieve a particular objective. The problem is, like so many other issues, the great communicator Barack Obama has not been really successful in selling the Iran nuclear treaty to the American people as a big success.
So my advice to President Obama: Forget conventional campaigning. It never works; it never helps. It's never made a difference, since 1860, in the history of the American presidency, and concentrate on selling your accomplishments to the American people, particularly the Iran nuclear treaty. The administration could also achieve a big success if it really makes enormous headway in the conflict against Isis [another name for the Islamic State]. It's made some headway, and there's a real chance ISIS will be driven back from most of its territory before the presidential election. It could also achieve a success by a real cease-fire in Syria. These are all hanging fire. So that key is still up in the air.
Q: Which keys do you think might decide this election?
LICHTMAN: Right now the Democrats are clearly down four keys: The mandate key, which is based on midterm elections - they got shellacked in 2014. The sitting president key. Obviously Obama isn't running again. The policy change key, which only applies to the second term, and there's been no significant, lasting policy changes in the second term. And the incumbent charisma key. Whatever you may think of Hillary Clinton, she's no John F. Kennedy or Franklin Roosevelt.
That means this election will be decided by the two keys hanging fire: The contest key, No. 2, and the foreign policy success key, No. 11. It will not be decided by the debates, the speeches, the ads, the tricks of the campaign.
And finally, there is also this kind of wild-card factor of, will the Republican Party split? And will there be a Republican candidate who is an alternative to Donald Trump?
Q: So what do you say to people who argue, "Well, the polls don't match up with what he's saying?"
LICHTMAN: The polls are often wrong. Let me give you the example of 1988. I predicted the 1988 election in May of 1988, when George H.W. Bush was 17 points behind Michael Dukakis in the polls. I wrote in Washingtonian magazine: "Forget the polls, forget the pundits. Not only is George Bush going to win, he is a shoo-in to win. Because as weak a candidate as he is, he's running on the Ronald Reagan record of peace and prosperity, and the Democrats can't beat that, particularly with a conventional candidate like Michael Dukakis." So if you're going to go by the early polls, you'll be right most of the time, but you'll also be wrong a significant part of the time. Remember, the polls are only a current snapshot. They are not predictive.
Of course, you journalists don't pay attention to that and treat the horse-race polls as though they were predictive of what's going to happen many months from now. They're not. Whereas the keys to the White House, which don't use any polls, have two advantages. No. 1, they're based upon a theory of how elections really work, not just on random looking at polls. No. 2, they tell you exactly what has to happen for one party or the other party to win. And thirdly, very often, they can make predictions years ahead of time. This is a difficult election to call, which is why I haven't called it yet, but it's not an impossible election to call.
- The Washington Post