MIDTERM congressional elections nearly always mean a loss of seats in the House and Senate for members of a president's party. Since 1950, the president's party has lost an average of 24 seats in the House and three seats in the Senate during midterm elections. Not since the midterm elections in 1934 has a president's party actually gained seats in both the House and Senate.
Charged with obstructing President Franklin D. Roosevelt's New Deal agenda, Republicans lost nine seats in the House and 10 seats in the Senate, giving Democrats super-majorities in both chambers -
an electoral thrashing Haley Barbour, Sen. Bob Dole (R) of Kansas, and Rep. Newt Gingrich (R) of Georgia would be wise to recall before counting their expected gains next month. Democrats will lose a fair number of seats in Congress in November. The question is: How many? One indicator used by politicos and pundits is the president's public-approval rating: As goes Clinton, so go the Democrats. With Clinton's public-approval rating hovering at about 39 percent - the lowest such rating prior to a midterm election since President Truman in 1950 - Republicans are confidently predicting a rout of the Democrats on election day.
In a fund-raising letter sent out by the National Republican Congressional Committee in August, Mr. Gingrich claims, ``The Republican Party has a very real chance to win a working Republican majority in Congress for the first time in 40 years. Out of 259 House seats the Democrats will be defending, all we need to do is win 25 of those races in order to gain working control of the House.'' A recent plea for contributions to fund the National Republican Senatorial Committee's ``voter turn-out plan'' by Mr. Dole suggests that Republicans are ``either leading or within striking distance of winning in 14 out of the 22 Democrat seats up for reelection.'' Republicans need only seven seats to take majority control of the Senate for the first time since 1986.
GOP optimism is certainly warranted. However, American electoral history suggests that Republicans had better not stake their expectations for control of either congressional chamber on Clinton's low public approval rating - at least not exclusively.
In House elections there is some evidence to substantiate a connection between a president's public-approval ratings and seats lost by the president's party. But there's little evidence of the same effect in Senate elections.
Since the midterm election of 1950, the five presidents with the lowest public approval ratings prior to a November midterm election also lost the five largest number of House seats for their party. These five presidents, with an average public approval rating of 45 percent, lost an average of 40 House seats for their party.
However, ``low public approval'' is a relative phrase. The Democrats lost 29 House seats in 1950 when Truman's approval rating sank to 32 percent. Eight years later, with President Eisenhower's public approval at 52 percent, Republicans lost 48 House seats. A low public approval rating for the president adds to the losses experienced by the party in power during a midterm election. But, it does not predict the magnitude of the loss.
Had Clinton brought Democrats into the House of Representatives on his coattails in 1992, Republicans might reasonably expect party turnover in the House to be very high, in the 30- to 40-seat range. But Democrats actually lost nine House seats in 1992, hence some of the turnover that might have been ``due'' the GOP already has taken place. House turnover is likely to fall short of the 42 seats necessary to comprise a new Republican majority. Nevertheless, Republicans could secure the 25 seats necessary for ``a working majority'' in the House in concert with Southern Democrats.
Senate elections are different, as the Founders intended them to be. There is only a scant relationship between presidential popularity and Senate seats won or lost by the party in power. Democrats lost six Senate seats in 1950 when Truman's approval was down to 32 percent, but Republicans actually gained one Senate seat in 1982 when Reagan's popularity dipped to 43 percent before the midterm election. In fact, since 1950, the two largest midterm losses in the Senate for the president's party occurred when the president's popularity was above 50 percent. Republicans lost 13 Senate seats in 1958 even though Eisenhower had a 52 percent public approval rating. Likewise, with Reagan's approval rating at 63 percent - the second highest of any post-war president prior to the midterm elections - Republicans still lost 8 Senate seats and control of the Senate.
Consequently, presidential popularity is a rather weak indicator of how many Senate seats a president's party may win or lose in midterm elections - weak enough to make predictions based on Clinton's current popularity very undependable. For Republicans to capture control of the Senate, they need to win more than twice the number of seats lost on average by the president's party in midterm elections - possible, but not likely enough for Mr. Dole to order new stationery.
The Opinion/Essay Page welcomes manuscripts. Authors of articles we accept will be notified by telephone. Authors of articles not accepted will be notified by postcard. Send manuscripts by mail to Opinions/Essays, One Norway Street, Boston, MA 02115, by fax to 617 -450-2317, or by Internet E-mail to OPED@RACHEL.CSPS.COM.