This paper updates through the 2000 election the equation originally presented in Fair (1978) explaining votes for president. Previous updates are in Fair ( 1982, 1988, 1990, 1996, 1998). The specification of the equation has not been changed from the previous update, which itself was unchanged from the update before that. The equation has simply been reestimated using the latest data. The data are discussed in Section II; the estimates are presented in Section III; and predictions for 2004 are presented in Section IV. The appendix contains a complete description of how the data were collected and a listing of all the data. The appendix allows one to duplicate the results in this paper if desired.
A non technical discussion of the vote equation is contained in Fair (2002), Predicting Presidential Elections and Other Things .
Since the last update the Department of Commerce has revised the national income and product data back to 1929, and the revised data have been used for the current update. Data prior to 1929 have been obtained, as before, from Balke and Gordon (1986). The appendix discusses the splicing of the Balke and Gordon data to the Department of Commerce data.
The notation for the variables is as follows:
The equation has VOTE on the left hand side and the other variables plus a constant term on the right hand side. It is linear in coefficients. The sample period begins with the 1916 election. The equation is estimated by ordinary least squares.
Column 1 of Table 1 below presents the estimates using the entire 1916-2000 sample period (22 observations). This is the equation that is used for the predictions in the next section. It is the "updated" equation. Column 2 presents for comparison purposes the previous estimates, which are based on the 1916-1996 sample period and the pre revised data. These are the estimates presented in Box 3-2, p. 47, in Fair (2002). Column 3 presents the estimates using the revised data but only the 1916-1960 sample period (12 observations).
The predicted values and errors are presented at the end of the table for each of the three sets of estimates. For column 2 the predicted value for 2000 is outside sample. For column 3 the predicted values for 1964 and beyond are outside sample, where the predicted value for 2000 is outside sample by 40 years! The predictions for column 2 use the pre revised data except for the prediction for 2000, which uses the updated data. Columns 1 and 3 use the updated data exclusively.
__________________________________________________________________________
Table 1
Three Sets of Estimates of the Vote Equation
1 2 3
Previous
Sample: 1916-2000 1916-1996 1916-1960
GROWTH .691 .698 .805
(6.72) (7.45) (7.94)
INFLATION -.775 -.720 -.477
(-2.71) (-2.77) (-1.34)
GOODNEWS .837 .905 .701
(3.12) (3.86) (2.90)
PERSON 3.25 3.99 5.21
(2.50) (3.25) (4.26)
DURATION -3.63 -3.33 -2.08
(-3.04) (-3.07) (-2.34)
PARTY -2.71 -2.84 -3.58
(-4.65) (-5.21) (-6.23)
WAR 3.85 4.65 3.88
(1.46) (1.96) (1.72)
INTERCEPT 49.61 48.40 47.36
(18.08) (19.10) (21.88)
SE 0.0237 0.0215 0.0147
2
R 0.923 0.941 0.987
No. obs. 22 21 12
V | V | V | V
Year act. | pred. error | pred. error | pred. error
| | |
1916 51.7 | 50.9 -0.8 | 50.7 -0.9 | 50.9 -0.8
1920 36.1 | 39.2 3.1 | 38.9 2.8 | 36.4 0.2
1924 58.2 | 57.3 -1.0 | 57.9 -0.4 | 57.6 -0.7
1928 58.8 | 57.6 -1.2 | 57.3 -1.5 | 57.4 -1.4
1932 40.8 | 38.8 -2.1 | 39.2 -1.7 | 41.2 0.4
1936 62.5 | 63.8 1.4 | 64.3 1.8 | 63.5 1.1
1940 55.0 | 55.7 0.7 | 56.0 1.0 | 55.4 0.4
1944 53.8 | 52.5 -1.2 | 52.9 -0.9 | 53.8 0.1
1948 52.4 | 50.5 -1.8 | 50.5 -1.9 | 52.1 -0.3
1952 44.6 | 44.4 -0.2 | 43.9 -0.7 | 43.9 -0.7
1956 57.8 | 57.3 -0.5 | 57.3 -0.5 | 57.6 -0.2
1960 49.9 | 51.6 1.7 | 51.1 1.2 | 51.8 1.9
1964 61.3 | 61.1 -0.3 | 61.2 -0.1 | 59.5 -1.8
1968 49.6 | 50.2 0.6 | 49.6 0.0 | 49.2 -0.4
1972 61.8 | 59.4 -2.4 | 59.8 -2.0 | 61.6 -0.2
1976 48.9 | 48.9 0.0 | 48.6 -0.4 | 51.3 2.3
1980 44.7 | 45.7 1.0 | 45.6 0.9 | 45.8 1.1
1984 59.2 | 62.0 2.9 | 61.4 2.3 | 63.7 4.6
1988 53.9 | 51.3 -2.6 | 52.4 -1.5 | 52.1 -1.8
1992 46.5 | 51.7 5.1 | 50.9 4.4 | 55.2 8.6
1996 54.7 | 53.7 -1.0 | 52.7 -2.1 | 53.0 -1.8
2000 50.3 | 48.9 -1.3 | 48.5 -1.8 | 47.1 -3.2
_____________________________________________________________________________
Comparing columns 1 and 2, the coefficient estimates are fairly close. The updating has resulted in no major changes. This is as expected because the previous equation (column 2) predicted the 2000 election well (an error of -1.8 percentage points), and so it is unlikely that adding this observation and reestimating would make much difference. A remarkable feature of the results is how well the equation estimated only through 1960 does (column 3). For this equation the eight coefficient estimates are based on only twelve observations. There are ten outside sample predictions for this equation, beginning with 1964, and the mean absolute error is only 2.6 percentage points for these ten predictions. As noted above, the 2000 prediction is outside sample by 40 years.
The predictions show that the most problematic election for the equation is 1992, where President Bush was predicted to win and yet lost to Clinton by a fairly large amount. Much of the discussion in Fair (1996) is concerned with trying to account for this error. The results for 1996 and 2000, on the other hand, are good, with errors in column 1 of only -1.0 and -1.3 percentage points respectively.
The updated coefficient estimate for GROWTH is .691, which says that a one percentage point increase in GROWTH leads to a .691 percentage point increase in the vote share. The coefficient estimate for INFLATION is -.775, which says that a one percentage point increase in INFLATION leads to a .775 percentage point decrease in the vote share. The coefficient estimate for GOODNEWS is .837, which says that each good news quarter adds .837 percentage points to the vote share.
For the 2004 election PARTY is -1, DURATION is 0, and WAR is 0. If President Bush runs again, which is assumed here, then PERSON is 1. Multiplying these values by their respective coefficients and adding the intercept gives a value of 55.57. The vote equation is then:
The October 31, 2002, economic forecast from the US model on this site has no more predicted good news quarters, so GOODNEWS has a predicted value of 1. GROWTH is predicted to be 2.0 percent (assuming population growth of 0.9). INFLATION is predicted to be 1.9 percent, which is based on the past seven inflation rates and the eight predicted future rates. Using these economic forecasts in the above equation gives a predicted value of VOTE of 56.3 percent. President Bush is thus predicted to win by a fairly confortable margin. This is essentially the same result that was reached in Fair (2002) using the previous vote equation (see Box 4-2, p. 65).
The equation is telling a fairly simple story. President Bush has a head start because he is a Republican and is the incumbent running again. In addition, the duration variable is not working against him because the Republicans have only been in power for one consecutive term. Regarding the economy, inflation is predicted to be low, and this is good for Bush. The predicted growth rate in the year of the election is modest, which is more or less neutral for Bush, and the predicted number of good news quarters is very low, which is bad for Bush. All this adds up to a predicted value for Bush of about 56 percent of the two-party vote.
It would take a fairly poor economy for the equation to predict a Bush loss. For example, if inflation were 5 percent and the growth rate in the year of the election -4 percent, the predicted vote share would be 49.8 percent, a narrow Bush loss. Remember, however, that the estimated standard error is 2.37 percentage points, and so if the predicted vote share gets within about 3 percentage points of one half, there is considerably uncertainty regarding the winner.
The data that were used to estimate the vote equation are presented in Table A-1 below. Data are presented back to 1880 even though only data back to 1916 were used in the estimation work. Quarterly data on nominal GDP, real GDP, and population are needed to construct GROWTH, INFLATION, and GOODNEWS, and these data are presented in Table A-2 at the end of this paper. In Table A-2 Y denotes nominal GDP, X denotes real GDP, and POP denotes population. Let a subscript k denote the kth quarter of the sixteen-quarter period of an administration. Also, let Z = X/POP, per capita real GDP, and let P = Y/X, the GDP deflator. Then GROWTH and INFLATION are constructed as:
VOTE is the incumbent vote divided by the Democratic plus Republican vote except for the 1912 and 1924 elections. For 1912, VOTE is the incumbent vote divided by the Democratic plus Republican plus Roosevelt vote. For 1924, VOTE is the incumbent vote plus .765 times the LaFollette vote divided by the Democratic plus Republican plus LaFollette vote. The voting data for 1880-1916 were obtained from U.S. Department of Commerce (1975), pp. 1078-1079. For 1920-1932 the data were obtained from U.S. Department of Commerce (1988), p. 232, for 1936-1992 the data were obtained from U.S. Department of Commerce (1997), p. 271, and for 1996-2000 the data were obtained from U.S. Department of Commerce (2001), p. 233.
PARTY, PERSON, DURATION, and WAR are defined in the text. In the construction of PERSON Ford is not counted as an incumbent running again, since he was not an elected vice president, whereas the other vice presidents who became president while in office are counted.
The data on nominal GDP were obtained as follows. Annual data for 1929-1945 and quarterly data for 1946:1-2002:3 were obtained from the Bureau of Economic Analysis (BEA) website on October 31, 2002. Quarterly data for 1877:1-1945:4 are available from Balke and Gordon (1986), pp. 789-795. The Balke and Gordon values for 1877:1-1928:4 were used exactly, but the values for 1929:1-1945:4 were adjusted to take account of the new BEA annual data. For 1929:1-1945:4 each quarterly value for a given year was multipled by a splicing factor for that year. The splicing factor is the ratio of the BEA value for that year to the respective yearly value in Balke and Gordon (1976), pp. 782-783.
The data on real GDP were obtained in a similar way. Annual data for 1929-1946 and quarterly data for 1947:1-2002:3 were obtained from the BEA website on October 31, 2002. Quarterly data for 1877:1-1946:4 are available from Balke and Gordon (1986), pp. 789-795. The Balke and Gordon values were spliced to the BEA values. All the Balke and Gordon quarterly values for 1877:1-1929:4 were multiplied by the same number. This number is the ratio of the BEA value for 1929 to the 1929 value in Balke and Gordon (1976), p. 782. For 1930:1-1946:4 each Balke and Gordon quarterly value for a given year was multipled by a splicing factor for that year. The splicing factor is the ratio of the BEA value for that year to the respective yearly value in Balke and Gordon (1976), pp. 782-783.
The data on population were obtained as follows. For 1877-1928 annual data were obtained from U.S. Department of Commerce (1973), pp. 200-201, A114 series. Each of these observations was multiplied by 1.000887, a splicing factor. The splicing factor is the ratio of the A114 value for 1929 in U.S. Department of Commerce (1973) to the value for 1929 in Table 8.2 in U.S. Department of Commerce (1992). For 1929-1945 annual data were obtained from U.S. Department of Commerce (1992), Table 8.2. Quarterly observations for 1877:1-1945:4 were obtained by interpolating the annual observations using the method presented in Fair (1994), Table B.6. For 1946:1-2002:3 quarterly data were obtained from the BEA website on October 31, 2002.
_____________________________________________________________
Table A-1
Data Used in the Estimation
VOTE PARTY PERSON DURATION WAR GROWTH INFLATION GOODNEWS
1880 50.220 -1 0 1.75 0 3.879 1.974 9
1884 49.846 -1 0 2.00 0 1.589 1.055 2
1888 50.414 1 1 0.00 0 -5.553 0.604 3
1892 48.268 -1 1 0.00 0 2.763 2.274 7
1896 47.760 1 0 0.00 0 -10.024 3.410 6
1900 53.171 -1 1 0.00 0 -1.425 2.548 7
1904 60.006 -1 0 1.00 0 -2.421 1.442 5
1908 54.483 -1 0 1.25 0 -6.281 1.879 8
1912 54.708 -1 1 1.50 0 4.164 2.172 8
1916 51.682 1 1 0.00 0 2.229 4.252 3
1920 36.119 1 0 1.00 1 -11.463 0.000 0
1924 58.244 -1 1 0.00 0 -3.872 5.161 10
1928 58.820 -1 0 1.00 0 4.623 0.183 7
1932 40.841 -1 1 1.25 0 -14.557 7.160 4
1936 62.458 1 1 0.00 0 11.677 2.454 9
1940 54.999 1 1 1.00 0 3.611 0.055 8
1944 53.774 1 1 1.25 1 4.433 0.000 0
1948 52.370 1 1 1.50 1 2.858 0.000 0
1952 44.595 1 0 1.75 0 0.840 2.316 6
1956 57.764 -1 1 0.00 0 -1.394 1.930 5
1960 49.913 -1 0 1.00 0 0.417 1.963 5
1964 61.344 1 1 0.00 0 5.109 1.267 10
1968 49.596 1 0 1.00 0 5.070 3.156 7
1972 61.789 -1 1 0.00 0 6.125 4.813 4
1976 48.948 -1 0 1.00 0 4.026 7.579 4
1980 44.697 1 1 0.00 0 -3.594 7.926 5
1984 59.170 -1 1 0.00 0 5.568 5.286 8
1988 53.902 -1 0 1.00 0 2.261 3.001 4
1992 46.545 -1 1 1.25 0 2.223 3.333 2
1996 54.736 1 1 0.00 0 2.712 2.146 4
2000 50.265 1 0 1.00 0 1.603 1.679 7
______________________________________________________________