The Myth of America's Turn to the Right

But is this really where the voters are? Has the American public shifted to the right? In view of recent legislation and executive action it is hard to argue with the proposition that a right turn in American public policy has indeed taken place. Following some much less significant precedents established by Democratic and Republican Administrations in the 1970s, the Reagan Administration has broken sharply with many of the major policies championed by New Deal-oriented Democratic (and even Republican) Presidents from the 1930s through the mid-1960s. Its current tax-reform proposals aside, it has passed extremely regressive tax legislation. It has cut social welfare programs and has made plain its desire for far deeper cuts. It has dismantled or gutted a host of regulatory programs and abandoned the view that the federal government has a major positive role to play in guaranteeing the rights of minorities and of the victims of discrimination. More than any other Administration since the 1930s, it has moved away from the New Deal tradition of multilateral internationalism and free trade. While shifting dollars away from social programs, it has sponsored the largest sustained peacetime military buildup in U.S. history. And it has vastly increased the size of federal budget deficits, thus creating system-wide pressures for a reduction in the scope of federal activity.

Together, the policy initiatives of the Reagan Administration announce the end of the New Deal era in American politics. Because the Democratic Party originated that era, and set the agenda for national politics during it, these initiatives also announce the decline of the Democratic Party as the dominant force in American public life. What might be termed the New Deal party system, within which the Democrats held the upper hand, has finally collapsed.

But what is at stake in the current policy changes in American politics is of concern to more than Democratic loyalists. The very structure of American politics has changed, with the center of gravity of the American party system--including both Democrats and Republicans--shifting to the right. Because the stakes are so high, it is important to consider closely the basic assumptions of political debate now shared by the two major parties. Above all, it is vital to know if the central claim made by revisionist Democrats and Republicans alike--that a majority of the public has reached a stable, well-informed consensus on the desirability of right, or center-right, policies--is true.

We do not believe that it is. While there is overwhelming evidence of a policy realignment, there is little direct evidence that mass public sentiment has turned against the domestic programs of the New Deal, or even the most important components of the Great Society, and little evidence of a stable shift to the right in public attitudes on military and foreign policy. On the contrary, poll after poll demonstrates that the basic structure of public opinion in the United States has remained relatively stable in recent years. To the extent that there have been changes in public opinion on particular issues, most have tended to run against the direction of public policy. Moreover, despite major Republican victories in 1980 and 1984, recent voting behavior and trends in partisan identification provide little evidence of an electoral realignment.

American public opinion has long been best described as both ideologically conservative and programmatically liberal. That is, Americans are opposed to big government, and respond favorably to the myths and symbols of competitive capitalism in the abstract. When it comes to assessing specific government programs or the behavior of actual business enterprises, however, they support government spending in a variety of domestic areas and are profoundly suspicious of big business. Similarly, Americans are strongly anti-Communist and generally hostile to the Soviet Union, but they are wary of using force in the pursuit of U.S. foreign-policy objectives and anxious to live in peace with Russia. This basic opinion structure may appear schizophrenic, but as Walter Dean Burnham, a political scientist at the Massachusetts Institute of Technology, has observed, at least on domestic issues it probably reflects voter adaptation to a political system in which the government is interventionist but nonsocialist. In any case, it has been stable for at least a generation.

Within this structure, moreover, the trend in public opinion over the past generation has been toward greater liberalism. During the 1970s, and particularly after 1973, the rate of increase in support for liberalism slowed somewhat. In addition, as we will note below, there were some exceptions to the general liberal trend. But none of this amounted to a change in the basically liberal direction of public opinion.

We begin with domestic issues. Revisionist Democrats commonly claim that voter support for their party's traditional positions on typical New Deal issues--including government management of the economy; protecting the Social Security system, the aged, and workers; and, at least since the mid-1960s, medical care--has long been in decline, and that the present weakness of the party at the national level is a long-term consequence of that fact. Typically it is argued that rising incomes in the post-New Deal era have made voters more middle class, both in the sense that the majority of Americans are living far better than their parents did (and therefore no longer support social programs) and in the sense that they no longer respond positively to FDR-style attacks on "economic royalists," made in the name of the common man.

The survey data, however, do not bear this argument out. A recent effort by the Princeton political scientist Stanley Kelley to map the "salience" (that is, the importance to voters in their choice of party or candidate) of such New Deal issues found a remarkable stability in voter attitudes in presidential elections from 1952 to 1976. In 1952, 85 percent of the voters found New Deal issues salient--that is, 85 percent of the electorate cited positions on such issues as something to like or dislike about the major parties or their candidates or both. The figure dropped to 76 percent of the electorate in 1956, and varied only slightly over the next five elections, at 74 to 77 percent of the electorate. Obviously, New Deal issues have not become less important to American voters since the late 1950s. Moreover, among those who saw a position on New Deal issues as "biasing" their choice--that is, leading them to favor a particular party or candidate--the Democrats enjoyed the support of strong majorities in every election during the 1952-1976 period. The average Democratic bias was just under 62 percent. Put otherwise, if those elections had been decided solely on those issues, Democrats would have won them all by landslides. Clearly, Democratic Party identification with traditional positions on New Deal issues was a major party strength, not a weakness.

The stability of such attitudes aside, it is commonly argued that the public moved sharply to the right during the late 1970s on most domestic issues, and that a groundswell of popular opposition to domestic-spending programs and government regulation of business was an important reason for Ronald Reagan's election in 1980. In fact no such groundswell occurred. To the contrary, public skepticism toward business, and support for government regulation of it, actually increased on several dimensions during the 1970s. From 1969 to 1979, for example, the share of the public thinking that there was "too much power concentrated in the hands of a few large companies for the good of the nation" increased from 61 to 79 percent; those thinking that business as a whole was making "too much profit" grew from 38 to 51 percent. Over the period 1971 to 1979 the percentage thinking that "government should put a limit on the profits companies can make" nearly doubled, rising from 33 to 60 percent.

Nor was there any evidence of an upsurge in support for domestic-spending cuts. The National Opinion Research Center found in 1980 that only 21 percent of Americans (the average across the different spending areas) thought that "too much" was being spent on environmental health, education, welfare, and urban-aid programs--the same percentage holding that belief in 1976, 1977, and 1978. The percentage of Americans who thought that "too little" was spent on those programs was also remarkably stable over the 1976-1980 period, dropping only from 44 to 42 percent, while the combined percentage of those who thought the amount spent was "too little" or "about right" was never lower than 72 percent.

As the rollback in regulation and cutbacks in domestic spending became evident during Reagan's first term, the public increased its support for regulatory and social programs. A *Los Angeles Times* poll in 1982, for example, asked respondents if they would favor "keeping" or "easing" regulations that President Reagan "[said] are holding back American free enterprise." Even given this language, the percentage favoring "keeping" outweighed that favoring "easing" regulations regarding the environment (49-28), industrial safety (66-18), the teenage minimum wage (58-29), auto emission and safety standards (59-29), federal lands (43-27), and offshore oil drilling (46-29). In 1983 the *Los Angeles Times* again asked about regulatory policy and found that only five percent of Americans considered regulations "too strict," while 42 percent thought they were "not strong enough." A CBS News/*New York Times* poll, meanwhile, found that support for the relatively extreme proposition that "protecting the environment is so important that requirements and standards cannot be too high, and continuing environmental improvements must be made regardless of cost" increased from 45 percent in 1981 to 58 percent in 1983.

The movement in support of social-spending programs was similar. An NBC News poll found that those agreeing with the statement that Reagan was "going too far in attempting to cut back or eliminate government social programs" rose from 37 to 52 percent over 1981-1983. Comparing the results of a 1982 survey with one commissioned in 1978, the Chicago Council on Foreign Relations found a remarkable 26-percentage-point jump in those wishing to "expand" rather than "cut black" welfare and relief programs, typically the easiest target of attack among social-spending programs. Support for government action to assist the unemployed was particularly broad. By early 1983 a CBS News/*New York Times* poll found that 74 percent of the public favored a government jobs program even if it meant increasing the federal deficit.

Such sentiments were neither swamped by the Reagan recovery nor abandoned in the 1984 landslide for the incumbent President. Even on the eve of Reagan's second inauguration, with his approval rating at a record 68 percent and drums beating loudly about the huge federal deficit, a *Washington Post*/ABC News Poll found that only 35 percent of Americans favored substantial cuts in social programs to reduce the deficit (although 65 percent believed such cuts were on the way). Representative was a CBS/*New York Times* poll conducted shortly after the election which compared responses then with those in a post-election survey in 1980:

"There is no suggestion in this poll that the American public has grown more conservative during the four years of the Reagan administration. If anything, there is more willingness now to spend money on domestic programs and a better assessment of the social programs of the 1960's."

A year later the results were similar. A January, 1986, CBS/*New York Times* survey found that voter attitudes showed "no consistent evidence of change, certainly not in a conservative direction," over the course of the Reagan presidency. On social programs, for example, the poll showed that:

"Fewer people (40 percent versus 51 percent as Mr. Reagan was coming into office) think that welfare recipients could get along without welfare payments. Fewer people (51 percent versus 63) believe that government creates more problems than it solves....Even more significantly, perhaps, fully 66 percent think the Government should spend money now on efforts similar to those of the Great Society programs to help the poor people in the United States. That finding seems to run counter to many Democrats' view that there is no longer any political mileage in such Federal poverty programs."

On social issues as well, American attitudes betray no evidence of a right turn. In general, religion, feminism, civil liberties, abortion, and race relations are the policy areas in which the public has shown the sharpest increase in liberalism since the Second World War. The rate of increase has slowed during the post-1973 period, but at no time has the public actually become more conservative on these issues. And on several of the key questions highlighted by the Reagan Administration the public became more liberal over the course of Reagan's first term. Regarding abortion, for example, an NBC News exit poll in 1984 found that two thirds of the electorate endorsed the legalization of abortion, with the decision "left to the woman and her physician," while only a quarter did not. Soon after the election an ABC News poll found that the share of Americans supporting the relatively radical position that women should have a right to abortion on demand, "no matter what the reason," actually increased over Reagan's first term, rising from 40 percent in 1981 to 52 percent in 1985, while the percentage opposing abortion on demand declined from 59 to 46 percent. According to a *Los Angeles Times* poll, only 23 percent of the electorate supported a constitutional amendment prohibiting abortion, and only 32 percent of those who voted for Reagan endorsed his policy on abortion. Louis Harris reports that Americans oppose requiring school prayer by 51 to 43 percent and support passage of the Equal Rights Amendment by 60 to 34 percent. And few Americans are enthusiasts of Jerry Falwell, the most prominent leader of the religious right. A 1984 *Los Angeles Times* exit poll found that only 16 percent of the voters approved of the minister.

Recent trends in public attitudes toward affirmative action merit special notice, if only because of the Reagan Administration's sustained attack on affirmative action and the revisionists' repeated insistence on the public's impatience with such "special interests" as women and minorities. In 1978, Louis Harris reported, only 45 percent of the public agreed that "if there are no affirmative action programs helping women and minorities in employment and education, then these groups will continue to fail to get their share of jobs and higher education, thereby continuing past discrimination in the future"; 36 percent disagreed. By 1982 the percentage agreeing had risen to 57, and 39 percent disagreed; by September of 1985, 71 percent agreed and 27 percent disagreed. Those favoring a "federal law requiring affirmative action programs for women and minorities in employment and education, provided there are no rigid quotas," numbered 67 percent by 1984, and those opposed, 18 percent; by 1985 support had risen to 75 percent (opposition was 21 percent). Here, too, the movement of public opinion has been directly opposite to the movement of public policy.

Once again, in noting the generally liberal trend in American public opinion on domestic issues in recent years, we are not arguing that this trend is universal. American support for harsher sentencing of criminals has grown dramatically over the past two decades, and this may be interpreted as a move toward greater conservatism. Resistance to tax increases--although, significantly, not demands for tax reduction--grew during the 1970s, and very likely worked to soften support for increased social spending, at least in some areas. Given all the countervailing trends, however, it would obviously be mistaken to generalize from these two exceptions, and considered on their own terms, neither advances the revisionist case.

Support for harsher sentencing of criminals began to grow in the mid-1960s. The trend coincides with the real and dramatic increase in crime observable since then, and not with the much later right turn in public policy. Nor has growing support for harsher sentences interrupted trends of increasing support for social tolerance and civil liberties, and declining support for violent responses to crime (questions of perceived self-defense aside). It thus seems to reflect public frustration with a manifest policy failure, not a more general lurch to the right.

Attitudes toward taxes, on the other hand, are surprisingly muted and lenient, given the background facts. Over the past generation the burden and unfairness of the tax system has increased substantially for Americans of average income. From the mid-1950s to 1980 the tax burden of the average American family nearly doubled, and from the mid-1960s to 1980 the federal tax system became significantly less progressive, as the corporate share of federal tax revenues dropped sharply and the share provided by regressive payroll taxes skyrocketed. It is not surprising that by the 1970s overwhelming numbers of Americans thought the tax system was unfair. In 1977, near the high point of the much-publicized "tax revolt," Harris reported that 89 percent of Americans thought that "the big tax burden falls on the little man in this country"; in 1978 Harris reported that 80 percent thought taxes were "unreasonable." The same year, H&R Block reported that 74 percent believed middle-income families were paying too much in taxes, while 76 percent thought high-income families were paying too little, and 72 percent thought the same about large corporations.

Despite this very widespread perception of unfairness, however, the *increase* in the (always large) share of the population that thought its own taxes were too high was negligible during the 1970s. In Harris polls it rose only one percentage point over the 1969-1978 period. Public opinion on trading off services and taxes was also stable in the late 1970s. The percentage who favored keeping "taxes and services about where they are" (45) was the same in 1975 and 1980; so was the percentage (38) favoring a decrease in services and taxes; the small share who wished to increase both actually rose slightly over the period. Finally, it is important to emphasize that among the general public (as opposed to business elites) the salience of the tax issue has never been great; only tiny percentages have ever named "high taxes" as the country's most important problem; and the 1970s were no exception.

On balance, then, it seems reasonable to conclude that the heavier burdens on average Americans, and especially the greater unfairness in the tax system, made the public more resistant to tax increases and more receptive to promises to cut taxes. This did not, however, set off a groundswell of public clamor for the reduction of taxes, and it certainly did not provide a mandate for the regressive revisions of the tax system made during Reagan's first term. As of January, 1985, three quarters of Americans thought the tax system was "unfair to the ordinary working man or woman." Here again, then, movements in public opinion have a highly imperfect, or even negative, relation to movements in public policy.

And then there is foreign policy. Here, should be said, there is evidence, particularly in the late 1970s and early 1980s, of a conservative turn in public attitudes on the important issue of military spending. According to the Gallup Poll, from 1976 to 1981 the percentage of Americans who believed that the United States spent too much on defense declined from 36 to 15 percent, while the share that believed the United States spent too little grew from 22 to 51 percent. This shift is by far the strongest piece of evidence for the claim that American public opinion has moved to the right, and it is commonly cited as a mandate, delivered to and acted upon by both the Carter and the Reagan Administration, to increase U.S. military outlays sharply. That it is the only case of strong convergence between public views and public-policy changes should give pause to those making such a claim. There are other reasons, as well, why this mandate is not clearcut evidence for the revisionist Democrats' case.

First, the largest part of the shift occurred very late in the period, in 1979 and 1980, after military authorizations had already begun rising dramatically. As late as December of 1978 NBC News/Associated Press showed that 50 percent of the population actually favored a decrease in military spending over either an increase or a continuation of spending at current levels. Since popular consensus on the need for a military buildup only followed that buildup, it cannot be readily identified as its cause.

Second, even admitting this exceptional congruence between public opinion and policy change, there is good reason to believe that the shift in public opinion was itself produced by elite action. At the risk of belaboring the obvious, we should point out that the public does not form its opinions in a vacuum, and by the late 1970s anxious elites were mobilizing an enormous campaign to increase military spending. Careful studies of media views of military spending, for example, show increasingly hawkish sentiments expressed over the period and show the public's opinions consistently following, not leading, the views of elites. This campaign reached a crescendo in 1979-1980, when the fall of the Pahlevi regime in Iran, the seizure of the American embassy in Tehran, and the Soviet invasion of Afghanistan were repeatedly cited as proof that U.S. military capabilities were seriously weakened. By 1980, with both parties calling for increases in military spending, with Carter, the Democratic candidate, denouncing the Soviet invasion of Afghanistan as "the worst threat to world peace since World War Two," and with Reagan, the Republican candidate, insisting that the United States had "unilaterally disarmed" during the previous decade, opposition to a military buildup simply ceased to be a respectable political position. In this context it is not surprising that large sections of the public expressed concern about U.S. military readiness. But this picture contrasts sharply with familiar claims that the public simply "woke up" to the need for a military buildup.

Third, the military "mandate," as it turned out, was fragile and short-lived. After peaking in 1980 and early 1981, support for increasing military outlays dropped sharply. Harris reported that the percentage of Americans in favor of "increasing...the present defense budget" fell from 71 percent in 1980 to 14 percent by early 1983. An ABC/*Washington Post* poll found that those who thought the Reagan buildup was "going too far" rose from 28 percent in 1981 to 51 percent in 1983, while the National Opinion Research Center found that those who favored reductions in military spending rose from 11 to 32 percent over 1980-1983. By that time, indeed, the general distribution of attitudes on military spending resembled the situation in the early 1970s, close to the height of domestic opposition to the Vietnam War.

Fourth, and finally, the strength of public support for any given spending initiative is best assessed in the context of other spending priorities. Public perception of the need for trade-offs between military and social outlays clearly played an important role in diminishing support for the buildup, and does not help the revisionist case. ABC News/*Washington Post* polls over the period from early 1981 to late 1983 showed that support for increased domestic spending grew from 49 to 67 percent for programs directed to the poor, from 43 to 75 percent for education programs, and from 49 to 66 percent for health programs--all while support for military spending declined. The Harris Survey provided even more direct evidence on this point by asking respondents to choose between cutting back particular social programs and cutting back military outlays. For federal aid to cities those preferring cuts in military spending to cuts in the program grew from 41 to 54 percent over 1981-1984, for federal aid to education from 60 to 71 percent, for federal health programs from 58 to 67 percent. For unemployment compensation they grew from 46 to 70 percent over 1981-1982 alone. Here, too, it is very difficult to reconcile changes in public opinion with the changes in public policy inaugurated during Reagan's first term.

For all these reasons, then, the increased support for military spending is a weak peg on which to rest the revisionist case.

Even granting a brief increase in support for additional military spending, however, what about public support for the use of force abroad? Once again there is a split between Americans' professed ideological positions and their willingness to support specific, especially costly, acts of intervention. As Richard Lugar, the Republican chairman of the Senate Foreign Relations Committee, summarized this state of affairs in January, 1985:

"In poll after poll, Americans express their concern about hostile governments which imperiled our interests in Latin America and elsewhere. But in the same polls, Americans display an equal and overwhelming opposition to any course of action which might actually frustrate governments which are harmful to us."

More concretely, while short-term military operations against tiny opponents are sometimes (briefly) applauded, as in the case of the invasion of Grenada, Americans are generally opposed to sustained military interventions abroad. During Reagan's first term early majority support in 1982 for a U.S. Marine presence in Lebanon turned to solid opposition by late 1983. By September, Harris reported 61 percent disapproving of Reagan's handling of the situation there. The bombing of the Marine barracks in late October produced a brief boost in public support for a continued military presence--Gallup reported that it increased from 28 to 34 percent between October and late November--but by mid-January of 1984 it was back down to 28 percent. Disapproval of Reagan's general handling of the situation rose from 52 percent in November of 1983 to 60 percent by early February of 1984.

With regard to other U.S. interventions similar results were reported. Over the period 1982-1984 Harris reported general disapproval of Administration policies in El Salvador consistently running at around 65 percent. Opposition to sending troops to El Salvador was even higher: on major network surveys it averaged more than 80 percent. Government policies toward Nicaragua were also unpopular. Opposition to the U.S. mining of Nicaragua's harbors, according to a CBS/*New York Times* poll, ran at 67 percent; the same poll reported two-to-one majorities opposed to aid to the contras, even when the question was, Should the United States "help people in Nicaragua who are trying to overthrow the pro-Soviet government there"?

Finally, in what is arguably the single most important area of foreign and military policy--U.S. strategic policy toward, and relations with, the Soviet Union--recent polls do indicate a profound shift in public sentiment. Here again, however, the change runs in a direction exactly opposite to the direction of policy change during Reagan's first term. Despite the talk among some Administration officials of winnable nuclear war, and Reagan's own bellicose rhetoric against the "evil force," survey data indicate that public aversion to nuclear weapons is strong and growing, that arms control meets with wide public approval, and that an overwhelming majority of Americans, while continuing to disapprove of the Soviet Union, are more reluctant than ever to prompt a showdown with the USSR and want to live more peacefully with it. A late 1984 survey of public attitudes toward the USSR, co-authored by the noted pollster Daniel Yankelovich, summed up the American mood as one of "live-and-let-live pragmatism." Two thirds of Americans believe that the United States should let the Soviets have their system of government while we retain ours, since "there's room in the world for both," and 53 percent think the United States would actually be safer if it stopped trying to halt the spread of communism to other countries (only 22 percent disagreed).

Fear of nuclear war and widespread skepticism about the possibility of "winning" the arms race provide the foundations for this new pragmatism. Ninety-six percent of Americans believe that "picking a fight with the Soviet Union is too dangerous in a nuclear world"; 89 percent believe "there can be no winner in an all-out nuclear war...both the United States and the Soviet Union would be completely destroyed"; 68 percent (and 78 percent of those under thirty) think that if both sides keep building nuclear weapons, instead of negotiating their elimination, it is only a matter of time before they are used; and 38 percent (50 percent of those under thirty) think that a full-scale nuclear war is likely to occur in the next ten years. In light of these views it is striking that 50 percent of Americans think that the Reagan Administration is playing nuclear "chicken" with the Soviet Union.

Regarding arms, Americans certainly do not want to "lose" the arms race (57 percent, for example, think the United States should continue to build new and "better" nuclear weapons), but equally certainly they do not think it can be "won." Eighty-four percent of Americans believe that the United States no longer has nuclear superiority; 90 percent think that both the United States and the Soviet Union already have an "overkill" capacity; and 92 percent believe that if the United States developed a bigger nuclear arsenal than the Soviets', the USSR would simply keep building until it matched the United States.

Indeed, concern about nuclear weapons and the current state of U.S.-Soviet relations leads strong majorities of Americans to favor substantial revisions of U.S nuclear strategy and positions on arms control: 77 percent believe that the United States should soon have a policy of not using nuclear weapons to respond to a conventional Soviet attack; 74 percent believe that the United States should now have a policy of never using nuclear weapons in a battlefield situation; upwards of 75 percent have for years favored a bilateral nuclear freeze; 61 percent favor a *unilateral* six-month freeze on nuclear-weapons development to see if the Soviets respond in kind, even if they take advantage of the gesture; 56 percent support signing an arms-control agreement even if foolproof verification is impossible; and 84 percent are opposed to using new weapons as bargaining chips in arms-control negotiations.

But if public opinion has generally run against the right turn in public policy why was Reagan so spectacularly popular during his first term, and especially during the first two years of that term, when most of his Administration's major programs were put in place? The answer is simple: he was not.

Press claims that Reagan's first term marked him as the most popular President of the postwar period are simply mistaken. Gallup reports that over the length of that term his approval rating--which measures satisfaction with job performance--averaged 50 percent, lower than the averages of Eisenhower (69), Kennedy (71), Johnson (52), and Nixon (56), and not far above that of Carter (47). Reagan's high point was reached shortly after taking office, when, in May of 1981, he recorded a 68 percent approval rating. Four out of the five other elected postwar Presidents had higher peaks--Eisenhower (79 percent), Kennedy (83), Johnson (80), and Carter (75)--while Nixon's 67 percent approval peak was not significantly lower. In addition, Reagan's low point of 35 percent, reached in January, 1983, was significantly lower than those of Eisenhower and Kennedy, matched the level that Johnson sank to just before he withdrew from national politics, in 1968, and came close to the 24 percent approval rating that Nixon received in July and August of 1974, just before he resigned the presidency in disgrace.

During the first two years of Reagan's presidency, when his Administration made its greatest advances, his approval ratings were particularly unimpressive. The average approval ratings for his first and second year were 54 and 44 percent, respectively. The first- and second-year average percent approval ratings for Carter (62 and 47), Nixon (61 and 57), Kennedy (75 and 72), and Eisenhower (69 and 65) were all substantially higher.

As Reagan's approval rating fell in the polls over the 1981-1982 period, the press began emphasizing his enormous personal popularity. Readers and viewers were told that although the public did not like what Reagan was doing to them, they loved the man who was doing it. There is some measure of truth in this. Reagan's ratings for personal appeal have always remained well above his ratings for job performance. But press accounts commonly did not go on to point out that significant differentials between performance- and personal-approval ratings of Presidents are utterly routine, that they always show greater personal approval than performance approval (Americans, for whatever reason, want to believe that their Presidents are nice people), and that in fact the differential Reagan enjoyed was proportionately *smaller* than those of his postwar predecessors, not larger.

Of course, Reagan's approval ratings rose spectacularly over the 1983-1984 period, climbing from their low of 35 percent in early 1983 to a strong, though hardly unprecedented, 61 percent by the time of the election. (Reagan's average approval ratings for his third and fourth years were 44 and 56 percent, respectively, as compared with Carter's 38 and 40 percent, Nixon's 50 and 56 percent, Eisenhower's 71 and 73 percent, and Kennedy's 1963 rating of 64 percent.) They have continued strong ever since. But as virtually all careful analyses of the trends in Reagan's approval ratings conclude, this upward swing, like the downward one before it, seems to have been caused almost entirely by changes in economic conditions in the country (see the first chart on page 50). It gives no evidence of a magic bond between Reagan and the public. Indeed, if one controls for economic conditions, Reagan's popularity does not differ significantly from Jimmy Carter's.

Polls leading up to the election also fail to support the myth of the Great Persuader. ABC News/*Washington Post* surveys, for example, asked voters if they liked the President personally, irrespective of his policies, and then asked if they intended to vote for him. What emerged consistently from these polls was that personal approval was far less important than policy approval in determining voter choice. For example, a survey in May of 1984 reported that 70 percent of those who said they liked Reagan personally but disapproved of his policies were not planning to vote for him.

Finally, it is important to stress that the policy approval Reagan did receive in the 1984 vote was almost entirely confined to judgments on recent economic performance and on hopes that the economy would continue to boom along. With economic considerations swamping other policy concerns, this led to anomalous results: vast numbers of voters cast ballots for Reagan even though they disagreed with his positions on a host of issues. This was particularly striking among voters who thought they would be better off under a Reagan second term. A *Washington Post*/ABC News poll in September of 1984 found that 53 percent of registered voters thought they would be personally better off. Many of this group disagreed with Reagan on a range of issues; nonetheless, they said overwhelmingly that they would vote for him (see the second chart on page 50).

Exit polls, as already indicated, confirmed such results. On virtually all the important issues identified with the "Reagan revolution" in public policy, public opinion ran against the President. Moreover, of those voting for Reagan only six percent (as compared with an already low 11 percent in 1980) directly identified his conservatism as one of the things that mattered to them, and only five percent offered agreement with his views as their chief reason for support.

The conclusion is inescapable. With the exception of the rise in support for increased military spending, which was rapidly reversed, there is little or nothing in the public-opinion data to support the claim that the American public moved to the right in the years preceding Reagan's 1980 victory. If American public opinion drifted anywhere over Reagan's first term, it was toward the left. And there is nothing in the data to support the claim that Reagan's first term marked him as the most popular President in modern times or that he has a magical hold over the electorate.

Most studies of American voting behavior also fail to find evidence of a major realignment of voter allegiance toward the Republicans. The most carefully conducted studies of "policy voting" in 1980, for example, conclude that the vote was more a negative referendum on Carter than an endorsement of Reagan's policies.

In 1984, to be sure, Reagan's personal victory was spectacular. He took 59.2 percent of the two-party vote and scored the second greatest electoral-college landslide in modern American history (exceeded only by Roosevelt's percentage in 1936). Below the presidential level, however, the gains were unimpressive. In 1980, in a less sweeping national victory, Reagan brought thirty-three new representatives and twelve new senators with him. In 1984, in contrast, the Democrats posted a net loss of only fourteen seats in the House and a net gain of two seats in the Senate. In state elections they dropped one governorship and control over five of the ninety-nine state legislative houses. These minor losses did not erase the gains made during the 1982 midterm elections. Indeed, the Republicans now have slightly fewer seats in the Senate and the House, and half a dozen fewer governorships, than when Reagan first took office.

If anything, the election gives evidence of further electoral *dealignment*, with voting defined ever less sharply along partisan lines, and a continued decline in the capacity of conventional politics to organize and integrate electoral demand. On such basic dimensions of electoral vitality as overall participation rates and competition in congressional races, for example, the election results again suggested continued decay. As a percentage of the potential electorate, voter turnout was stagnant, rising only 0.7 percent from its 1980 level to run at 55 percent. Some 77 million eligible Americans--24 million more than voted for Reagan--abstained. Thus while Reagan's 59.2 percent share of the two-party vote was a certifiable landslide, his share of the potential electorate, as in 1980, was certifiably not. In 1980 Reagan garnered only 27.6 percent of the potential electorate, a smaller share than Wendell Wilkie won in his decisive 1940 loss to Roosevelt. In 1984 Reagan garnered 32.3 percent, or a smaller share than that gained by Eisenhower in both his elections, by either Kennedy or Nixon in 1960, by Johnson in 1964, or by Nixon in 1972.

Congressional competition reached an all-time low in 1984. Both the number (408) and the percentage (93.8) of congressional races including incumbents reached historic highs, as did the number (392) and percentage (96.1) of races won by incumbents. There were also relatively few close races. A stunning 74.9 percent of winning candidates took 60 percent or more of the vote--the highest such figure on record, with records going back to 1824. More immediately, as the divergent results at the presidential and sub-national level reported above indicate, there was very little partisan channeling of the vote that Reagan got. Despite sharp differences in the general political climate in 1982 and 1984, in the 340 House races where candidates from both parties faced each other both times, the net swing toward the Republican Party over the 1982-1984 period was a paltry 3.7 percentage points.

Some analysts of recent trends in party identification nevertheless point to a rise in the percentage of Americans, and particularly in the percentage of young people, identifying with the Republican Party. At first glance, again, the evidence looks highly suggestive. In the last quarter of 1984 Gallup and other polls reported that the ratio of Democratic "identifiers" to Republican ones was about 1.1: 1, closer to statistical parity than at any time in the previous forty years. And a 15-percentage-point swing toward the Republicans among younger (eighteen-to-twenty-nine-year-old) voters in the 1984 election led all other age groups. Unfortunately for the revisionists, however, this proves next to nothing.

First, public affection for the two parties is volatile. Public opinion on which party is best able to handle "the most important problem" facing the country, for example, which is commonly used as an indicator of public support for each party, has zigzagged repeatedly over the past four presidential elections. Explicit identification with the major parties was less volatile in the late 1970s but has become more so in the past few years. Over the period 1975-1979 the Gallup Poll reported that Republican identifiers composed 21 to 23 percent of the electorate, with Democratic identifiers taking a 45 to 48 percent share. Then, as the economy declined under Jimmy Carter, Reagan won the 1980 election, and the Republicans moved aggressively in 1981 on their tax and spending proposals, the number of Republican identifiers increased dramatically. It rose from 21 percent of the electorate in the first quarter of 1980 to 28 percent in the third quarter of 1981; over the same period Democratic identifiers dropped from 47 to 41 percent. At the time, there was excited talk of a realignment in American politics. But just as quickly as the Republicans had narrowed the gap to thirteen points, it widened once again. Over the next seven quarters Republican identifiers fell to 23 percent of the electorate and Democratic identifiers increased to 46 percent. Finally, as the recovery took hold, the Democrats ran a campaign that promised higher taxes but not economic growth, and Reagan won another election, the Republicans surged forward again in a party-identification split of 38 percent Democratic, 35 percent Republican in the fourth quarter of 1984.

Of course, such short-term volatility in party identification, which suggests the need for caution in interpreting the figures, could be consistent with a long-term trend toward the Republican Party. However, the data on party identification over time do not support this claim either.

The virtual parity between the major parties reported in late 1984 and 1985, for example, is not unprecedented in the post-New Deal period. During the 1940s, when, revisionists claim, the "New Deal coalition of groups" was at its peak, comparable or even closer conditions of parity obtained. In 1940 and 1944 the ratios were the same as in late 1984 (1.1:1); in 1946 the parties were perfectly balanced at 1:1; after an uptick for the Democrats in the late 1940s they returned to roughly the same level (1.2:1) in 1952, and bounced around considerably afterward. Given the ups and downs in the ratio, if one takes the whole post-New Deal period as a point of reference, different trends in party identification can be shown, depending on what year one begins with. If one begins in 1937, 1949, or 1952, the trend (albeit slight) in party identification has actually been toward the Democrats. If one begins in 1964, the high point of Democratic identification, the trend (again slight) has been toward the Republicans. Since the data can be used to show different things, the real question is where it is most appropriate to begin the observations. The data alone, of course, do not answer that.

The volatility and ambiguity of trends suggest deeper problems in such identification data. Party identification is interesting chiefly as a predictor of how people will vote once they get inside a polling booth and how they can be expected to continue to vote in the future. But both the election results reported earlier and the volatility and inconclusiveness of trends in party sentiment emphasized here give reason to doubt the usefulness of identification data in making such predictions. The absence of partisan channeling of the vote in 1984, after all, was not a new phenomenon. It was only a very striking case of what has long been observed of American voters--that they seem to resort less and less to parties as a key to their vote at all.

While direct evidence of this is hard to come by, it is suggestive that over the past five national elections the percentage of voters who reported having voted a "straight ticket" (all Republican or all Democratic) has been a consistent minority, ranging between 37 and 44 percent of the electorate. Also, the percentage of congressional districts choosing representatives of one party and a President of another has risen sharply since the mid-1960s, climbing from just over 26 percent in 1960 to close to 44 percent in 1984. Clearly, most voters do not feel bound by party labels. More and more, it seems, voter responses to questions about identification record little more than the party identity of the candidate they voted for in the last election or intend to vote for in the next.

Against the argument that party identification among the young has shown a sharp trend toward the Republicans, all of the above can be noted again, with the additional caveat that the political sentiments of young people seem even more volatile than those of the rest of the electorate. One reason that the swing toward the Republican Party among young voters looked so big in 1984 was that they were swinging back from being the age cohort that most strongly endorsed Jimmy Carter in the election of 1980. Recent research suggests that young people also voted disproportionately Republican in the 1920s, just prior to the upsurge in Democratic support during the New Deal. The voting behavior of young people thus does not appear to be a very good predictor of political realignment.

Finally, it seems likely that if the Republicans were making serious strides in identification among voters, weak Democratic identifiers would be likely, especially if specifically invited, to join the Republican Party. But a recent effort to persuade 100,000 Democrats to register Republican, based on just that premise, fizzled embarrassingly; the Republican National Committee had to cover its expensive failure with a blanket of juggled statistics. And, despite an enormous expenditure of resources, the results of Republican registration efforts among non-voters in 1983-1984 were not particularly impressive.

To maintain the view that Reagan's policies do not reflect a new popular consensus on policy, of course, one needs an alternative explanation of just how he has managed to win two landslide election victories. But this question has an entirely plausible answer that leads to none of the conclusions drawn by the revisionist Democrats: *the economy*.

We have already noted that Reagan's approval rating has moved in tandem with the economy, and in this he is not exceptional. A huge amount of research confirms that economic performance is an important factor in whether voters approve or disapprove of elected officials, not only in the United States but in most capitalist democracies. Indeed, quite reasonable predictions of Reagan's percentage of the two-party vote in both 1980 and 1984 can be derived from a simple model that considers only changes in real disposable income per capita shortly before an election and a measure of presidential popularity (which correlates closely with economic performance). One widely disseminated model of this type fits the 1984 case almost perfectly, predicting a Reagan percentage of 59.4 percent--only 0.2 percent off the actual result.

Such "political business cycle" models suggest that the recent problems of the Democrats are more cyclical than secular. Above all they are matters of timing. Jimmy Carter pumped up the economy in 1977 and at the beginning of 1978 but then drove it into recession just in time for the 1980 election. He thus became the first President since Herbert Hoover to run for reelection at a moment when the national income was actually shrinking, and the first elected incumbent since Hoover to lose a bid for re-election. Ronald Reagan did not make this mistake. His Administration chose to have a long and exceptionally deep recession early in its term and then (helped along by the collapse of OPEC and other factors) staged one of the greatest political business cycles in modern history, producing a 5.8 percent increase in real per capita disposable income in 1984--the largest election-year increase since 1936 (see the third chart on page 50). Though one can argue to infinity whether the last few percent of Reagan's vote was or was not produced by this boom, the point is that income growth of this size cannot fail to produce a smashing victory for an incumbent, no matter what the other party does. It is not a reason for the opposition to agonize.

That the economy was the key issue in 1980 and 1984 is also confirmed by survey data. As already noted, virtually all studies of the 1980 election see it as a referendum on Carter's bad economic performance. In 1984 every major election poll found economic issues dominant in voters' minds. More immediately, and of considerable relevance to current plans to move the party to the right, a lack of confidence in the Democrats' program for the economy was the single most important reason, at the level of voting behavior, for the Democrats' defeat.

Consider voter opinion on the "New Deal issues" cited earlier. Contrary to revisionist claims, such issues continued to preoccupy voters in the 1980 and 1984 elections. In 1980, Stanley Kelley reports, New Deal issues were salient among 76 percent of the electorate; in 1984 their salience actually increased to 84 percent. In 1980, however, the Democratic bias on the composite of New Deal issues fell to 47 percent, some 16 points below its 1976 level, and in 1984 it recovered only to 53 percent. Thus if those elections had turned solely on New Deal issues, the Democrats would have lost resoundingly in 1980 and won only a narrow victory in 1984.

This drop-off in Democratic bias on New Deal issues had nothing to do with changed public perceptions of Democratic positions on such issues as Social Security, worker protection, assistance to the aged, medical care, or labor unions. On all those issues, in fact, the Democratic bias actually *increased* from 1976 to 1984. What changed was voter perception of the Democrats as macroeconomic managers.

Of concern to voters in all presidential elections during the period 1952-1984 was the election's expected impact on the economy (including the impact on employment, wages, salaries, working conditions, and inflation). This classic New Deal issue had an average salience of 25 percent over the period 1952-1976, and in those years the Democratic bias on the issue averaged an impressive 72 percent. In 1980 and 1984, however, two things happened. First, the issue became even more salient, rising to 32 percent in 1980 and 33 percent in 1984. Second, and most dramatic, the Democratic bias plunged to 20 percent in 1980 and recovered only to 29 percent in 1984. This whopping 47 percent drop between the average bias during the period 1952-1976 and that during 1980-1984, more than any other factor, explains voter rejection of Democrats at the polls. Significantly, it coincided with the party's move away from its traditional commitment to promoting employment gains and growth.

Such data point up the implausibility of the revisionist case as an electoral strategy. They suggest, indeed, that a move to the right may be the *worst* possible electoral move the Democrats could make, and that the mistake the Democrats made in 1984 was not their alleged "reaching down"--that is, their efforts to expand the electorate through voter-registration drives--but the fact that they did not reach down nearly enough. On economic issues the Democrats offered voters almost nothing in 1984. Walter Mondale became the first Democratic nominee in many years to fail even to put forward a major jobs program. Nor did he couple his call for tax increases with a program of economic revitalization. There is no particular mystery as to why voters, faced with a choice between someone who promised them little besides a rise in taxes and a candidate who at least verbally championed economic growth and lower taxes, deserted Mondale for Reagan or declined to vote at all. As one looks to 1988, the data suggest that a revisionist Democrat even more grimly committed to lowering expectations will be abandoned by the voters in even greater numbers.

Recent trends in public attitudes toward affirmative action merit special notice, if only because of the Reagan Administration's sustained attack on affirmative action and the revisionists' repeated insistence on the public's impatience with such "special interests" as women and minorities. In 1978, Louis Harris reported, only 45 percent of the public agreed that "if there are no affirmative action programs helping women and minorities in employment and education, then these groups will continue to fail to get their share of jobs and higher education, thereby continuing past discrimination in the future"; 36 percent disagreed. By 1982 the percentage agreeing had risen to 57, and 39 percent disagreed; by September of 1985, 71 percent agreed and 27 percent disagreed. Those favoring a "federal law requiring affirmative action programs for women and minorities in employment and education, provided there are no rigid quotas," numbered 67 percent by 1984, and those opposed, 18 percent; by 1985 support had risen to 75 percent (opposition was 21 percent). Here, too, the movement of public opinion has been directly opposite to the movement of public policy.

Once again, in noting the generally liberal trend in American public opinion on domestic issues in recent years, we are not arguing that this trend is universal. American support for harsher sentencing of criminals has grown dramatically over the past two decades, and this may be interpreted as a move toward greater conservatism. Resistance to tax increases--although, significantly, not demands for tax reduction--grew during the 1970s, and very likely worked to soften support for increased social spending, at least in some areas. Given all the countervailing trends, however, it would obviously be mistaken to generalize from these two exceptions, and considered on their own terms, neither advances the revisionist case.

Support for harsher sentencing of criminals began to grow in the mid-1960s. The trend coincides with the real and dramatic increase in crime observable since then, and not with the much later right turn in public policy. Nor has growing support for harsher sentences interrupted trends of increasing support for social tolerance and civil liberties, and declining support for violent responses to crime (questions of perceived self-defense aside). It thus seems to reflect public frustration with a manifest policy failure, not a more general lurch to the right.

Attitudes toward taxes, on the other hand, are surprisingly muted and lenient, given the background facts. Over the past generation the burden and unfairness of the tax system has increased substantially for Americans of average income. From the mid-1950s to 1980 the tax burden of the average American family nearly doubled, and from the mid-1960s to 1980 the federal tax system became significantly less progressive, as the corporate share of federal tax revenues dropped sharply and the share provided by regressive payroll taxes skyrocketed. It is not surprising that by the 1970s overwhelming numbers of Americans thought the tax system was unfair. In 1977, near the high point of the much-publicized "tax revolt," Harris reported that 89 percent of Americans thought that "the big tax burden falls on the little man in this country"; in 1978 Harris reported that 80 percent thought taxes were "unreasonable." The same year, H&R Block reported that 74 percent believed middle-income families were paying too much in taxes, while 76 percent thought high-income families were paying too little, and 72 percent thought the same about large corporations.

Despite this very widespread perception of unfairness, however, the *increase* in the (always large) share of the population that thought its own taxes were too high was negligible during the 1970s. In Harris polls it rose only one percentage point over the 1969-1978 period. Public opinion on trading off services and taxes was also stable in the late 1970s. The percentage who favored keeping "taxes and services about where they are" (45) was the same in 1975 and 1980; so was the percentage (38) favoring a decrease in services and taxes; the small share who wished to increase both actually rose slightly over the period. Finally, it is important to emphasize that among the general public (as opposed to business elites) the salience of the tax issue has never been great; only tiny percentages have ever named "high taxes" as the country's most important problem; and the 1970s were no exception.

On balance, then, it seems reasonable to conclude that the heavier burdens on average Americans, and especially the greater unfairness in the tax system, made the public more resistant to tax increases and more receptive to promises to cut taxes. This did not, however, set off a groundswell of public clamor for the reduction of taxes, and it certainly did not provide a mandate for the regressive revisions of the tax system made during Reagan's first term. As of January, 1985, three quarters of Americans thought the tax system was "unfair to the ordinary working man or woman." Here again, then, movements in public opinion have a highly imperfect, or even negative, relation to movements in public policy.

And then there is foreign policy. Here, should be said, there is evidence, particularly in the late 1970s and early 1980s, of a conservative turn in public attitudes on the important issue of military spending. According to the Gallup Poll, from 1976 to 1981 the percentage of Americans who believed that the United States spent too much on defense declined from 36 to 15 percent, while the share that believed the United States spent too little grew from 22 to 51 percent. This shift is by far the strongest piece of evidence for the claim that American public opinion has moved to the right, and it is commonly cited as a mandate, delivered to and acted upon by both the Carter and the Reagan Administration, to increase U.S. military outlays sharply. That it is the only case of strong convergence between public views and public-policy changes should give pause to those making such a claim. There are other reasons, as well, why this mandate is not clearcut evidence for the revisionist Democrats' case.

First, the largest part of the shift occurred very late in the period, in 1979 and 1980, after military authorizations had already begun rising dramatically. As late as December of 1978 NBC News/Associated Press showed that 50 percent of the population actually favored a decrease in military spending over either an increase or a continuation of spending at current levels. Since popular consensus on the need for a military buildup only followed that buildup, it cannot be readily identified as its cause.

Second, even admitting this exceptional congruence between public opinion and policy change, there is good reason to believe that the shift in public opinion was itself produced by elite action. At the risk of belaboring the obvious, we should point out that the public does not form its opinions in a vacuum, and by the late 1970s anxious elites were mobilizing an enormous campaign to increase military spending. Careful studies of media views of military spending, for example, show increasingly hawkish sentiments expressed over the period and show the public's opinions consistently following, not leading, the views of elites. This campaign reached a crescendo in 1979-1980, when the fall of the Pahlevi regime in Iran, the seizure of the American embassy in Tehran, and the Soviet invasion of Afghanistan were repeatedly cited as proof that U.S. military capabilities were seriously weakened. By 1980, with both parties calling for increases in military spending, with Carter, the Democratic candidate, denouncing the Soviet invasion of Afghanistan as "the worst threat to world peace since World War Two," and with Reagan, the Republican candidate, insisting that the United States had "unilaterally disarmed" during the previous decade, opposition to a military buildup simply ceased to be a respectable political position. In this context it is not surprising that large sections of the public expressed concern about U.S. military readiness. But this picture contrasts sharply with familiar claims that the public simply "woke up" to the need for a military buildup.

Third, the military "mandate," as it turned out, was fragile and short-lived. After peaking in 1980 and early 1981, support for increasing military outlays dropped sharply. Harris reported that the percentage of Americans in favor of "increasing...the present defense budget" fell from 71 percent in 1980 to 14 percent by early 1983. An ABC/*Washington Post* poll found that those who thought the Reagan buildup was "going too far" rose from 28 percent in 1981 to 51 percent in 1983, while the National Opinion Research Center found that those who favored reductions in military spending rose from 11 to 32 percent over 1980-1983. By that time, indeed, the general distribution of attitudes on military spending resembled the situation in the early 1970s, close to the height of domestic opposition to the Vietnam War.

Fourth, and finally, the strength of public support for any given spending initiative is best assessed in the context of other spending priorities. Public perception of the need for trade-offs between military and social outlays clearly played an important role in diminishing support for the buildup, and does not help the revisionist case. ABC News/*Washington Post* polls over the period from early 1981 to late 1983 showed that support for increased domestic spending grew from 49 to 67 percent for programs directed to the poor, from 43 to 75 percent for education programs, and from 49 to 66 percent for health programs--all while support for military spending declined. The Harris Survey provided even more direct evidence on this point by asking respondents to choose between cutting back particular social programs and cutting back military outlays. For federal aid to cities those preferring cuts in military spending to cuts in the program grew from 41 to 54 percent over 1981-1984, for federal aid to education from 60 to 71 percent, for federal health programs from 58 to 67 percent. For unemployment compensation they grew from 46 to 70 percent over 1981-1982 alone. Here, too, it is very difficult to reconcile changes in public opinion with the changes in public policy inaugurated during Reagan's first term.

For all these reasons, then, the increased support for military spending is a weak peg on which to rest the revisionist case.

Even granting a brief increase in support for additional military spending, however, what about public support for the use of force abroad? Once again there is a split between Americans' professed ideological positions and their willingness to support specific, especially costly, acts of intervention. As Richard Lugar, the Republican chairman of the Senate Foreign Relations Committee, summarized this state of affairs in January, 1985:

"In poll after poll, Americans express their concern about hostile governments which imperiled our interests in Latin America and elsewhere. But in the same polls, Americans display an equal and overwhelming opposition to any course of action which might actually frustrate governments which are harmful to us."

More concretely, while short-term military operations against tiny opponents are sometimes (briefly) applauded, as in the case of the invasion of Grenada, Americans are generally opposed to sustained military interventions abroad. During Reagan's first term early majority support in 1982 for a U.S. Marine presence in Lebanon turned to solid opposition by late 1983. By September, Harris reported 61 percent disapproving of Reagan's handling of the situation there. The bombing of the Marine barracks in late October produced a brief boost in public support for a continued military presence--Gallup reported that it increased from 28 to 34 percent between October and late November--but by mid-January of 1984 it was back down to 28 percent. Disapproval of Reagan's general handling of the situation rose from 52 percent in November of 1983 to 60 percent by early February of 1984.

With regard to other U.S. interventions similar results were reported. Over the period 1982-1984 Harris reported general disapproval of Administration policies in El Salvador consistently running at around 65 percent. Opposition to sending troops to El Salvador was even higher: on major network surveys it averaged more than 80 percent. Government policies toward Nicaragua were also unpopular. Opposition to the U.S. mining of Nicaragua's harbors, according to a CBS/*New York Times* poll, ran at 67 percent; the same poll reported two-to-one majorities opposed to aid to the contras, even when the question was, Should the United States "help people in Nicaragua who are trying to overthrow the pro-Soviet government there"?

Finally, in what is arguably the single most important area of foreign and military policy--U.S. strategic policy toward, and relations with, the Soviet Union--recent polls do indicate a profound shift in public sentiment. Here again, however, the change runs in a direction exactly opposite to the direction of policy change during Reagan's first term. Despite the talk among some Administration officials of winnable nuclear war, and Reagan's own bellicose rhetoric against the "evil force," survey data indicate that public aversion to nuclear weapons is strong and growing, that arms control meets with wide public approval, and that an overwhelming majority of Americans, while continuing to disapprove of the Soviet Union, are more reluctant than ever to prompt a showdown with the USSR and want to live more peacefully with it. A late 1984 survey of public attitudes toward the USSR, co-authored by the noted pollster Daniel Yankelovich, summed up the American mood as one of "live-and-let-live pragmatism." Two thirds of Americans believe that the United States should let the Soviets have their system of government while we retain ours, since "there's room in the world for both," and 53 percent think the United States would actually be safer if it stopped trying to halt the spread of communism to other countries (only 22 percent disagreed).

Fear of nuclear war and widespread skepticism about the possibility of "winning" the arms race provide the foundations for this new pragmatism. Ninety-six percent of Americans believe that "picking a fight with the Soviet Union is too dangerous in a nuclear world"; 89 percent believe "there can be no winner in an all-out nuclear war...both the United States and the Soviet Union would be completely destroyed"; 68 percent (and 78 percent of those under thirty) think that if both sides keep building nuclear weapons, instead of negotiating their elimination, it is only a matter of time before they are used; and 38 percent (50 percent of those under thirty) think that a full-scale nuclear war is likely to occur in the next ten years. In light of these views it is striking that 50 percent of Americans think that the Reagan Administration is playing nuclear "chicken" with the Soviet Union.

Regarding arms, Americans certainly do not want to "lose" the arms race (57 percent, for example, think the United States should continue to build new and "better" nuclear weapons), but equally certainly they do not think it can be "won." Eighty-four percent of Americans believe that the United States no longer has nuclear superiority; 90 percent think that both the United States and the Soviet Union already have an "overkill" capacity; and 92 percent believe that if the United States developed a bigger nuclear arsenal than the Soviets', the USSR would simply keep building until it matched the United States.

Indeed, concern about nuclear weapons and the current state of U.S.-Soviet relations leads strong majorities of Americans to favor substantial revisions of U.S nuclear strategy and positions on arms control: 77 percent believe that the United States should soon have a policy of not using nuclear weapons to respond to a conventional Soviet attack; 74 percent believe that the United States should now have a policy of never using nuclear weapons in a battlefield situation; upwards of 75 percent have for years favored a bilateral nuclear freeze; 61 percent favor a *unilateral* six-month freeze on nuclear-weapons development to see if the Soviets respond in kind, even if they take advantage of the gesture; 56 percent support signing an arms-control agreement even if foolproof verification is impossible; and 84 percent are opposed to using new weapons as bargaining chips in arms-control negotiations.

But if public opinion has generally run against the right turn in public policy why was Reagan so spectacularly popular during his first term, and especially during the first two years of that term, when most of his Administration's major programs were put in place? The answer is simple: he was not.

Press claims that Reagan's first term marked him as the most popular President of the postwar period are simply mistaken. Gallup reports that over the length of that term his approval rating--which measures satisfaction with job performance--averaged 50 percent, lower than the averages of Eisenhower (69), Kennedy (71), Johnson (52), and Nixon (56), and not far above that of Carter (47). Reagan's high point was reached shortly after taking office, when, in May of 1981, he recorded a 68 percent approval rating. Four out of the five other elected postwar Presidents had higher peaks--Eisenhower (79 percent), Kennedy (83), Johnson (80), and Carter (75)--while Nixon's 67 percent approval peak was not significantly lower. In addition, Reagan's low point of 35 percent, reached in January, 1983, was significantly lower than those of Eisenhower and Kennedy, matched the level that Johnson sank to just before he withdrew from national politics, in 1968, and came close to the 24 percent approval rating that Nixon received in July and August of 1974, just before he resigned the presidency in disgrace.

During the first two years of Reagan's presidency, when his Administration made its greatest advances, his approval ratings were particularly unimpressive. The average approval ratings for his first and second year were 54 and 44 percent, respectively. The first- and second-year average percent approval ratings for Carter (62 and 47), Nixon (61 and 57), Kennedy (75 and 72), and Eisenhower (69 and 65) were all substantially higher.

As Reagan's approval rating fell in the polls over the 1981-1982 period, the press began emphasizing his enormous personal popularity. Readers and viewers were told that although the public did not like what Reagan was doing to them, they loved the man who was doing it. There is some measure of truth in this. Reagan's ratings for personal appeal have always remained well above his ratings for job performance. But press accounts commonly did not go on to point out that significant differentials between performance- and personal-approval ratings of Presidents are utterly routine, that they always show greater personal approval than performance approval (Americans, for whatever reason, want to believe that their Presidents are nice people), and that in fact the differential Reagan enjoyed was proportionately *smaller* than those of his postwar predecessors, not larger.

Of course, Reagan's approval ratings rose spectacularly over the 1983-1984 period, climbing from their low of 35 percent in early 1983 to a strong, though hardly unprecedented, 61 percent by the time of the election. (Reagan's average approval ratings for his third and fourth years were 44 and 56 percent, respectively, as compared with Carter's 38 and 40 percent, Nixon's 50 and 56 percent, Eisenhower's 71 and 73 percent, and Kennedy's 1963 rating of 64 percent.) They have continued strong ever since. But as virtually all careful analyses of the trends in Reagan's approval ratings conclude, this upward swing, like the downward one before it, seems to have been caused almost entirely by changes in economic conditions in the country (see the first chart on page 50). It gives no evidence of a magic bond between Reagan and the public. Indeed, if one controls for economic conditions, Reagan's popularity does not differ significantly from Jimmy Carter's.

Polls leading up to the election also fail to support the myth of the Great Persuader. ABC News/*Washington Post* surveys, for example, asked voters if they liked the President personally, irrespective of his policies, and then asked if they intended to vote for him. What emerged consistently from these polls was that personal approval was far less important than policy approval in determining voter choice. For example, a survey in May of 1984 reported that 70 percent of those who said they liked Reagan personally but disapproved of his policies were not planning to vote for him.

Finally, it is important to stress that the policy approval Reagan did receive in the 1984 vote was almost entirely confined to judgments on recent economic performance and on hopes that the economy would continue to boom along. With economic considerations swamping other policy concerns, this led to anomalous results: vast numbers of voters cast ballots for Reagan even though they disagreed with his positions on a host of issues. This was particularly striking among voters who thought they would be better off under a Reagan second term. A *Washington Post*/ABC News poll in September of 1984 found that 53 percent of registered voters thought they would be personally better off. Many of this group disagreed with Reagan on a range of issues; nonetheless, they said overwhelmingly that they would vote for him (see the second chart on page 50).

Exit polls, as already indicated, confirmed such results. On virtually all the important issues identified with the "Reagan revolution" in public policy, public opinion ran against the President. Moreover, of those voting for Reagan only six percent (as compared with an already low 11 percent in 1980) directly identified his conservatism as one of the things that mattered to them, and only five percent offered agreement with his views as their chief reason for support.

The conclusion is inescapable. With the exception of the rise in support for increased military spending, which was rapidly reversed, there is little or nothing in the public-opinion data to support the claim that the American public moved to the right in the years preceding Reagan's 1980 victory. If American public opinion drifted anywhere over Reagan's first term, it was toward the left. And there is nothing in the data to support the claim that Reagan's first term marked him as the most popular President in modern times or that he has a magical hold over the electorate.

Most studies of American voting behavior also fail to find evidence of a major realignment of voter allegiance toward the Republicans. The most carefully conducted studies of "policy voting" in 1980, for example, conclude that the vote was more a negative referendum on Carter than an endorsement of Reagan's policies.

In 1984, to be sure, Reagan's personal victory was spectacular. He took 59.2 percent of the two-party vote and scored the second greatest electoral-college landslide in modern American history (exceeded only by Roosevelt's percentage in 1936). Below the presidential level, however, the gains were unimpressive. In 1980, in a less sweeping national victory, Reagan brought thirty-three new representatives and twelve new senators with him. In 1984, in contrast, the Democrats posted a net loss of only fourteen seats in the House and a net gain of two seats in the Senate. In state elections they dropped one governorship and control over five of the ninety-nine state legislative houses. These minor losses did not erase the gains made during the 1982 midterm elections. Indeed, the Republicans now have slightly fewer seats in the Senate and the House, and half a dozen fewer governorships, than when Reagan first took office.

If anything, the election gives evidence of further electoral *dealignment*, with voting defined ever less sharply along partisan lines, and a continued decline in the capacity of conventional politics to organize and integrate electoral demand. On such basic dimensions of electoral vitality as overall participation rates and competition in congressional races, for example, the election results again suggested continued decay. As a percentage of the potential electorate, voter turnout was stagnant, rising only 0.7 percent from its 1980 level to run at 55 percent. Some 77 million eligible Americans--24 million more than voted for Reagan--abstained. Thus while Reagan's 59.2 percent share of the two-party vote was a certifiable landslide, his share of the potential electorate, as in 1980, was certifiably not. In 1980 Reagan garnered only 27.6 percent of the potential electorate, a smaller share than Wendell Wilkie won in his decisive 1940 loss to Roosevelt. In 1984 Reagan garnered 32.3 percent, or a smaller share than that gained by Eisenhower in both his elections, by either Kennedy or Nixon in 1960, by Johnson in 1964, or by Nixon in 1972.

Congressional competition reached an all-time low in 1984. Both the number (408) and the percentage (93.8) of congressional races including incumbents reached historic highs, as did the number (392) and percentage (96.1) of races won by incumbents. There were also relatively few close races. A stunning 74.9 percent of winning candidates took 60 percent or more of the vote--the highest such figure on record, with records going back to 1824. More immediately, as the divergent results at the presidential and sub-national level reported above indicate, there was very little partisan channeling of the vote that Reagan got. Despite sharp differences in the general political climate in 1982 and 1984, in the 340 House races where candidates from both parties faced each other both times, the net swing toward the Republican Party over the 1982-1984 period was a paltry 3.7 percentage points.

Some analysts of recent trends in party identification nevertheless point to a rise in the percentage of Americans, and particularly in the percentage of young people, identifying with the Republican Party. At first glance, again, the evidence looks highly suggestive. In the last quarter of 1984 Gallup and other polls reported that the ratio of Democratic "identifiers" to Republican ones was about 1.1: 1, closer to statistical parity than at any time in the previous forty years. And a 15-percentage-point swing toward the Republicans among younger (eighteen-to-twenty-nine-year-old) voters in the 1984 election led all other age groups. Unfortunately for the revisionists, however, this proves next to nothing.

First, public affection for the two parties is volatile. Public opinion on which party is best able to handle "the most important problem" facing the country, for example, which is commonly used as an indicator of public support for each party, has zigzagged repeatedly over the past four presidential elections. Explicit identification with the major parties was less volatile in the late 1970s but has become more so in the past few years. Over the period 1975-1979 the Gallup Poll reported that Republican identifiers composed 21 to 23 percent of the electorate, with Democratic identifiers taking a 45 to 48 percent share. Then, as the economy declined under Jimmy Carter, Reagan won the 1980 election, and the Republicans moved aggressively in 1981 on their tax and spending proposals, the number of Republican identifiers increased dramatically. It rose from 21 percent of the electorate in the first quarter of 1980 to 28 percent in the third quarter of 1981; over the same period Democratic identifiers dropped from 47 to 41 percent. At the time, there was excited talk of a realignment in American politics. But just as quickly as the Republicans had narrowed the gap to thirteen points, it widened once again. Over the next seven quarters Republican identifiers fell to 23 percent of the electorate and Democratic identifiers increased to 46 percent. Finally, as the recovery took hold, the Democrats ran a campaign that promised higher taxes but not economic growth, and Reagan won another election, the Republicans surged forward again in a party-identification split of 38 percent Democratic, 35 percent Republican in the fourth quarter of 1984.

Of course, such short-term volatility in party identification, which suggests the need for caution in interpreting the figures, could be consistent with a long-term trend toward the Republican Party. However, the data on party identification over time do not support this claim either.

The virtual parity between the major parties reported in late 1984 and 1985, for example, is not unprecedented in the post-New Deal period. During the 1940s, when, revisionists claim, the "New Deal coalition of groups" was at its peak, comparable or even closer conditions of parity obtained. In 1940 and 1944 the ratios were the same as in late 1984 (1.1:1); in 1946 the parties were perfectly balanced at 1:1; after an uptick for the Democrats in the late 1940s they returned to roughly the same level (1.2:1) in 1952, and bounced around considerably afterward. Given the ups and downs in the ratio, if one takes the whole post-New Deal period as a point of reference, different trends in party identification can be shown, depending on what year one begins with. If one begins in 1937, 1949, or 1952, the trend (albeit slight) in party identification has actually been toward the Democrats. If one begins in 1964, the high point of Democratic identification, the trend (again slight) has been toward the Republicans. Since the data can be used to show different things, the real question is where it is most appropriate to begin the observations. The data alone, of course, do not answer that.

The volatility and ambiguity of trends suggest deeper problems in such identification data. Party identification is interesting chiefly as a predictor of how people will vote once they get inside a polling booth and how they can be expected to continue to vote in the future. But both the election results reported earlier and the volatility and inconclusiveness of trends in party sentiment emphasized here give reason to doubt the usefulness of identification data in making such predictions. The absence of partisan channeling of the vote in 1984, after all, was not a new phenomenon. It was only a very striking case of what has long been observed of American voters--that they seem to resort less and less to parties as a key to their vote at all.

While direct evidence of this is hard to come by, it is suggestive that over the past five national elections the percentage of voters who reported having voted a "straight ticket" (all Republican or all Democratic) has been a consistent minority, ranging between 37 and 44 percent of the electorate. Also, the percentage of congressional districts choosing representatives of one party and a President of another has risen sharply since the mid-1960s, climbing from just over 26 percent in 1960 to close to 44 percent in 1984. Clearly, most voters do not feel bound by party labels. More and more, it seems, voter responses to questions about identification record little more than the party identity of the candidate they voted for in the last election or intend to vote for in the next.

Against the argument that party identification among the young has shown a sharp trend toward the Republicans, all of the above can be noted again, with the additional caveat that the political sentiments of young people seem even more volatile than those of the rest of the electorate. One reason that the swing toward the Republican Party among young voters looked so big in 1984 was that they were swinging back from being the age cohort that most strongly endorsed Jimmy Carter in the election of 1980. Recent research suggests that young people also voted disproportionately Republican in the 1920s, just prior to the upsurge in Democratic support during the New Deal. The voting behavior of young people thus does not appear to be a very good predictor of political realignment.

Finally, it seems likely that if the Republicans were making serious strides in identification among voters, weak Democratic identifiers would be likely, especially if specifically invited, to join the Republican Party. But a recent effort to persuade 100,000 Democrats to register Republican, based on just that premise, fizzled embarrassingly; the Republican National Committee had to cover its expensive failure with a blanket of juggled statistics. And, despite an enormous expenditure of resources, the results of Republican registration efforts among non-voters in 1983-1984 were not particularly impressive.

To maintain the view that Reagan's policies do not reflect a new popular consensus on policy, of course, one needs an alternative explanation of just how he has managed to win two landslide election victories. But this question has an entirely plausible answer that leads to none of the conclusions drawn by the revisionist Democrats: *the economy*.

We have already noted that Reagan's approval rating has moved in tandem with the economy, and in this he is not exceptional. A huge amount of research confirms that economic performance is an important factor in whether voters approve or disapprove of elected officials, not only in the United States but in most capitalist democracies. Indeed, quite reasonable predictions of Reagan's percentage of the two-party vote in both 1980 and 1984 can be derived from a simple model that considers only changes in real disposable income per capita shortly before an election and a measure of presidential popularity (which correlates closely with economic performance). One widely disseminated model of this type fits the 1984 case almost perfectly, predicting a Reagan percentage of 59.4 percent--only 0.2 percent off the actual result.

Such "political business cycle" models suggest that the recent problems of the Democrats are more cyclical than secular. Above all they are matters of timing. Jimmy Carter pumped up the economy in 1977 and at the beginning of 1978 but then drove it into recession just in time for the 1980 election. He thus became the first President since Herbert Hoover to run for reelection at a moment when the national income was actually shrinking, and the first elected incumbent since Hoover to lose a bid for re-election. Ronald Reagan did not make this mistake. His Administration chose to have a long and exceptionally deep recession early in its term and then (helped along by the collapse of OPEC and other factors) staged one of the greatest political business cycles in modern history, producing a 5.8 percent increase in real per capita disposable income in 1984--the largest election-year increase since 1936 (see the third chart on page 50). Though one can argue to infinity whether the last few percent of Reagan's vote was or was not produced by this boom, the point is that income growth of this size cannot fail to produce a smashing victory for an incumbent, no matter what the other party does. It is not a reason for the opposition to agonize.

That the economy was the key issue in 1980 and 1984 is also confirmed by survey data. As already noted, virtually all studies of the 1980 election see it as a referendum on Carter's bad economic performance. In 1984 every major election poll found economic issues dominant in voters' minds. More immediately, and of considerable relevance to current plans to move the party to the right, a lack of confidence in the Democrats' program for the economy was the single most important reason, at the level of voting behavior, for the Democrats' defeat.

Consider voter opinion on the "New Deal issues" cited earlier. Contrary to revisionist claims, such issues continued to preoccupy voters in the 1980 and 1984 elections. In 1980, Stanley Kelley reports, New Deal issues were salient among 76 percent of the electorate; in 1984 their salience actually increased to 84 percent. In 1980, however, the Democratic bias on the composite of New Deal issues fell to 47 percent, some 16 points below its 1976 level, and in 1984 it recovered only to 53 percent. Thus if those elections had turned solely on New Deal issues, the Democrats would have lost resoundingly in 1980 and won only a narrow victory in 1984.

This drop-off in Democratic bias on New Deal issues had nothing to do with changed public perceptions of Democratic positions on such issues as Social Security, worker protection, assistance to the aged, medical care, or labor unions. On all those issues, in fact, the Democratic bias actually *increased* from 1976 to 1984. What changed was voter perception of the Democrats as macroeconomic managers.

Of concern to voters in all presidential elections during the period 1952-1984 was the election's expected impact on the economy (including the impact on employment, wages, salaries, working conditions, and inflation). This classic New Deal issue had an average salience of 25 percent over the period 1952-1976, and in those years the Democratic bias on the issue averaged an impressive 72 percent. In 1980 and 1984, however, two things happened. First, the issue became even more salient, rising to 32 percent in 1980 and 33 percent in 1984. Second, and most dramatic, the Democratic bias plunged to 20 percent in 1980 and recovered only to 29 percent in 1984. This whopping 47 percent drop between the average bias during the period 1952-1976 and that during 1980-1984, more than any other factor, explains voter rejection of Democrats at the polls. Significantly, it coincided with the party's move away from its traditional commitment to promoting employment gains and growth.

Such data point up the implausibility of the revisionist case as an electoral strategy. They suggest, indeed, that a move to the right may be the *worst* possible electoral move the Democrats could make, and that the mistake the Democrats made in 1984 was not their alleged "reaching down"--that is, their efforts to expand the electorate through voter-registration drives--but the fact that they did not reach down nearly enough. On economic issues the Democrats offered voters almost nothing in 1984. Walter Mondale became the first Democratic nominee in many years to fail even to put forward a major jobs program. Nor did he couple his call for tax increases with a program of economic revitalization. There is no particular mystery as to why voters, faced with a choice between someone who promised them little besides a rise in taxes and a candidate who at least verbally championed economic growth and lower taxes, deserted Mondale for Reagan or declined to vote at all. As one looks to 1988, the data suggest that a revisionist Democrat even more grimly committed to lowering expectations will be abandoned by the voters in even greater numbers.