Fifty economists were polled about 1992 by the Blue Chip Economic Indicators, a newsletter, early this month. Forty-nine predicted the economy will grow next year. One said it won't grow, but it won't decline, either.
That means, if Ray C. Fair knows what he's talking about, that George Bush will absolutely, positively be re-elected president in November.
Nothing that happens between now and then can change that. No matter who the Democrats nominate, no matter how well Pat Buchanan and David Duke do in the primaries, no matter what happens in the former Soviet Union or in Japan, no matter what Dan Quayle stumbles into, no matter what Saddam Hussein does or doesn't do, no matter how tongue-tied George Bush may be on the stump, no matter how incompetent his television commercials, no matter even how excellent my anti-Bush editorials and columns -- if the economy rises even a little or is flat, and if Ray Fair is right, George Bush wins.
And Ray Fair is usually right in his seemingly narrowly focused predictions. Mr. Fair is a Yale economist. Some years ago he developed a simple equation for predicting presidential elections. "The two economic variables that seem most important are the growth rate of real per capita GNP [gross national product] and the rate of inflation in the two-year period prior to the election," he explained. This is especially true when an incumbent president is running.
Professor Fair's equation has been tested for every election back to 1916. It correctly named the popular vote winner in 16 of those 19 elections. (It missed in 1960, 1968 and 1976, three close elections.) In July 1987, using Blue Chip Indicators economists' predictions about the 1988 economy, he forecast a Bush victory with 51.9 percent of the vote. Mr. Bush got 53.5.
Professor Fair sort of bases his precise vote prediction on a grid that relates inflation to GNP growth with Mr. Bush as the candidate. (See chart. Different percentages would be used if President Bush doesn't run -- about a 4 percent swing.) He said recently that unless the economy in 1992 suddenly comes to resemble that of 1980 (growth rate of -5.7 percent and inflation of 9 percent), President Bush will win easily. No one foresees such a development.
A number of academics play this predicting-by-the-numbers game. Professor Greg Markus of the University of Michigan (he's a political scientist, not an economist) developed a model in which the key components were measurements of individuals' personal economic situation and their estimate of the national economic situation.
In the last nine elections, Professor Markus' model has picked the winner far in advance six times, missing the percentage breakdown of the vote by an average of less than 3 percentage points in all nine. In the last two elections he's been off by only 0.1 percent and 0.67 percent.
Professor Markus says he foresees President Bush winning comfortably, if the consensus of predictions about the economy are even close to correct.
Another political scientist, Professor Alan Lichtman of American University, also believes you can predict the outcome of presidential elections with little concern about campaign events. However, he includes political and social indicators as well as economic ones.
His 1990 book, "The Thirteen Keys to the Presidency," includes assessments of the incumbency factor, the nature of a challenge, if any, within the party, "significant" third party activity, foreign policy and military successes and failures, scandal, social unrest, the charisma or lack thereof of the incumbent and of his general election opponent.
His formula has correctly "forecast" every presidential election since 1860, he demonstrates (tested ex post facto in most cases, of course, as were the economic models mentioned above).
If 8 of the 13 keys favor the president, he wins. Professor Lichtman says he can make no "flat prediction" about 1992 as of this moment. He gives Mr. Bush only 7 keys, with several undecided. He's waiting to see whether the Buchanan challenge or the David Duke third party challenge are serious enough to be a minus for Bush.
If neither is, and certain other unforeseen problems, such as massive rioting, for example, don't occur, then Mr. Lichtman will give President Bush enough keys to win victory.
I feel so certain that neither Mr. Buchanan nor Mr. Duke will be a factor that I give the president enough keys to forecast his victory now. In fact, by my assessment of events past, present and likely in the future, President Bush gets 8 keys even if Mr. Buchanan runs fairly well in the primaries and even if Mr. Duke runs as a third party candidate in November.
There is another way to forecast elections. It is less scientific, but it works. That is to look at the map. Except in unusual circumstances, Republicans always win presidential elections, because they have what Lyndon Johnson once called "the electoral lock."
He used that phrase in discussing presidential politics with an aide, Horace Busby. He said that as Republicans sewed up the South and West, Democrats could not win the presidency in the Electoral College even if they won popular vote victories based on good showings in the Northeast and Midwest.
The lock has gotten more pronounced, as Mr. Busby has pointed out on several occasions. The problem for the Democrats is that the Southern and Western states are growing in population, and thus in electoral votes, at the expense of the Northeast and Midwest states.
In the last six elections, the Democratic candidate has carried only this many of the 11 states of the old Confederacy and 18 states of the Plains, Mountain and Pacific West:
3 in 1968, 0 in 1972, 11 in 1976, 2 in 1980, 0 in 1984 and 3 in 1988. That is a cumulative total of 155 states to 19.
Only in 1976 did the Democrats win the White House. Mr. Busby says that was not a Democratic win but a Southern Baptist win. He means Jimmy Carter was supported in the South (but not the West) because of his religion and his social persona -- and despite his party and his politics. He won in the Midwest and Northeast because of his party and politics. Sounds reasonable. Nine of the 10 Southern states that he carried in 1976 went against him in 1980, when his party and politics were an issue again in the South.
I asked Mr. Busby if Bill Clinton of Arkansas could not duplicate Carter's 1976 performance. He said, and I quote, "no." I don't know why he is so sure. Mr. Clinton is a Baptist, but he went to Georgetown, Yale and Oxford. Southern conservatives don't relate to that sort of person the way they did to Mr. Carter, a Georgia Tech and Annapolis man.
I have a theory of my own that works pretty well in predicting presidents. It is that a man has to run -- or be sought -- for national office (president or vice president) and lose -- or decline -- before he can run for president and win. Going back to 1924, this theory worked in every election but 1976. Sixteen out of 17 ain't bad. It's not just a quirk. People don't like to vote for strangers when it comes to such an important office. Among Democrats, only Jerry Brown has run before. Even in the age of television, it takes a few years to become well known.
There are other historical clues that provide a distant early warning about who will win a presidential election. For example, no one has ever been elected from a state with single digit electoral votes since 1852.
Now that may be a quirk, a meaningless coincidence, but if it isn't, Governor Clinton of Arkansas (6 electoral votes), Sen. Tom Harkin of Iowa (7) and Sen. Bob Kerrey of Nebraska (5) -- one of whom is likely to be the Democratic nominee -- are doomed as challengers of George Bush.
There may be models -- economic, political, historical, statistical, trivial -- that forecast a Democratic victory next November, but if there are, I haven't seen them.