ELECTION WEATHER FORECAST
Virginia's Bellwethers May Point the Way to Victory and Defeat
By Larry J. Sabato
© Larry J. Sabato, UVA Center for Politics, 2001
So which way is the wind blowing in Virginia's 2001 governor's race? You can look to public opinion polls for the answer, though they often change with the wind. Or you can look to history.
There are "bellwether" cities and counties that always or almost always have voted with the winner of the Governor's Mansion in Virginia's modern two-party era, which began in 1969 with the election of the state's first Republican governor since the 1880s, Linwood Holton. Out of the 95 counties and 39 independent cities that currently exist in the Commonwealth, exactly TWO localities have gotten it right every single time.
Surprisingly, these localities are not found in the Old Dominion's heartland, but in the region considered -- at least by down-staters -- to be the most liberal and least representative, Northern Virginia. The small city of Fairfax and the gargantuan county of Fairfax have picked every governor in the eight elections between 1969 and 1997.
In addition to the twin Fairfaxes, there were 19 near-bellwethers. Five cities and fourteen counties chose the winner in seven of eight elections: the cities of Franklin, Galax, Lexington, Newport News, and Salem, and the counties of Bedford, Bland, Fluvanna, Gloucester, Lunenberg, Madison, Mathews, Middlesex, Montgomery, Orange, Prince Edward, Prince William, Southampton, and Wythe. In most cases, these localities have a closely split political personality. They are competitive with a substantial base of support for both Democrats and Republicans, yet neither party claims a solid majority of residents. Swing independents, moving from side to side with regularity, determine the identity of successful candidates there. Also, with the exceptions of Newport News city and Montgomery county, the near-bellwethers are not heavily populated, and they are geographically scattered.
At the other end of the spectrum are eight anti-bellwethers: Charles City, Greensville, New Kent, Norfolk, Petersburg, Portsmouth, Prince George, and Surry. These three cities and five counties have the state's worst record for picking winners -- just three of the last eight governors. With the exceptions of New Kent and Prince George counties, the anti-bellwethers are heavily Democratic with a large African-American constituency; naturally, they voted only for the three successful Democratic governors in the 1980s.
There is another way to measure bellwethers that is a little more precise. In addition to the number of prescient picks, we can also examine in each election year how close or far from the winner's statewide average vote each locality was. The accompanying list puts all of Virginia's localities in the order of most bellwether-like (Fairfax county) to least (Charles City county). Check and see how your locality fares in these rankings here.
How much stock should we place in bellwethers, especially as a predictive tool for 2001? The odds are substantial that, as a group, the bellwethers and near-bellwethers will probably be right again. Thus, the press and academics would be justified in sending reporters and researchers into these localities to test sentiments before the November vote.
However, consider this: before 1997, there were four, not two, perfect bellwethers. Even though Governor Jim Gilmore carried the vast majority of all the state's localities (106 of the then-135), he lost the small cities of Franklin and Lexington, which since 1969 had been totally on the money. Great confidence in any single bellwether, then, is misplaced. Even in the most accurate localities, the proportion of the votes given to the winning candidates have often not closely mirrored the statewide vote. For instance, Galax was 13.4% under the statewide average for Doug Wilder in 1989; Fairfax county voted 6.7% under the statewide average for George Allen in 1993; and Prince William was 7.2% over the average for Jim Gilmore in 1997.
Still, the crown belongs to "Fairfax squared" for now, and it is an impressive and highly selective title in Virginia. Our neighbor to the south, North Carolina, has many more bellwethers. Twenty-six counties there have picked the presidential winner consistently from 1968 to 2000, for example. (There are 100 counties in North Carolina, only a slightly larger number than in Virginia.)
Maybe November will prove that the crown rests uneasily on Virginia's northern champions. But for now, if you are heading to the voting booth or maybe even the racetrack, you might want to bring along a friend and adviser from Fairfax.
Sabato is the author of OVERTIME! THE 2000 ELECTION THRILLER (Longman), and he wishes to thank Joshua J. Scott, Peter Sheeran, and Cagri Tanyol of the UVa Center for Politics for their research assistance.