Presidential Scandals and Job Approval

Impact Analysis with SAS

by

Robert A. Yaffee

Statistics and Social Science Group
Academic Computing Facility
e-mail: robert.yaffee@nyu.edu

Keywords: Abuse of power, black-bag jobs, Clinton scandal, counterintelligence, corruption in government, coverup, covert action, electoral sabotage, hush money, Huston plan, impeachment, job approval, Lewinsky, Lewinsky scandal, McCarthyism, oath of office, obstruction of justice, perjury, President Clinton, public opinion, sale of ambassadorships, Watergate .

Social scientists frequently analyze the impact of events on a series of observations. Whether researchers are policy analysts, political scientists, economists, sociologists or engineers, they may wish to assess the effect of a discrete event or intervention on some measure of a process. Policy analysts might wish to study the impact of seat belt legislation or air bag legislation on the number of highway fatalities. Environmental scientists might wish to assess the effect of pollution control legislation on air and water quality. Research scientists may wish to understand the effect of ritalin on the perceptual speed of a particular patient. Political analysts need to assess the impact of a scandal on a president. Presidents depend on electoral support for continuance in office, until they have reached their last permitted term. A political crisis -- with its sudden appearance, great threat and limited time for response -- presents a challenge to the political efficacy and public image of the President. His success in responding to the threat to his administration can be measured by several operational indicators. Among the foremost is the extent to which the public approves of how he's doing his job. Another is the public impression of the President's political opposition. Is the political opposition poisoning the atmosphere with prejudicial leaks from unspecified sources to predispose the public and press toward a presumption of guilt, and if so, how frequently? Another indicator might be how many congressmen on the House Judiciary Committee will support him in an impeachment vote. How many congressmen will support him in a general House vote on the subject? How many senators will support him in a test of trial by the Senate? Another measure may be an overall assessment of his record prior to the eruption of the scandal. Another is how accurately and fairly the press reports the situation.

The Watergate scandal is one of the greatest political scandals in American political history. Although a proper analysis requires that each of many dimensions be analyzed, limitations of time and space force us to concentrate on only one of them here. Watergate, as we now know from the latest release of the Nixon White House Tapes, stemmed from the schemes, plans and covert operations of President Richard Nixon and select members of his staff. From these tapes we discover what Golda Meier observed, "As President Nixon says, presidents can do almost anything, and President Nixon has done many things that nobody would have thought of doing."

For purposes of illustration, we will focus on the assessment of Gallup Poll presidential job approval ratings. The question asked was, "Do you approve of the way the President is handling his job?" The answer categories are 1) Approve 2) Disapprove 3) No opinion. The percentage of the public approving is the measure under study. Data for this analysis have been culled from the Gallup Poll website when it was available for public access.
For a compilation of the data from several polls, the analyst may refer to the Roper Center at the University of Connecticut website at The Roper Center Presidential Job Approval Ratings. Graphical construction of this measure plotted over time helps put the events into perspective. For this purpose, SAS is the statistical package of choice because of its superior graphical capability, along with its excellent capability for modeling intervention analysis, compared to those of other packages. Several SAS graphs showing different phases of the Watergate scandal and subsequent impeachment crisis are shown in the figures.

In the Watergate scandal, there were so many significant events that it is necessary to break it down into phases. The background of the scandal is essential to understanding how it unfolded. New information was released in 1997, when Stanley Kutler's Abuse of Power was published. Kutler gained access to previously unreleased Nixon White House tapes through a Freedom of Information Act lawsuit. Kutler's transcripts of those tapes reveal startling new aspects of Nixon's covert campaign of political oppression. Nixon's secret plan to get the U.S. out of Vietnam in six months had evaporated, along with any remaining governmental credibility of success in the war. Public discontent was rife. Meanwhile, Nixon had become convinced that there was an anti-war political conspiracy out to get him (Kutler, pp. 15-17). On June 17, 1971, Nixon verbally re-endorsed the Huston Plan (named after aide Tom Charles Huston, who devised the scheme) for coordinated illegal espionage activities -- including burglaries, surreptitious entries, surreptitious surveillance, campus recruitment of informants and an array of assorted techniques to neutralize political opponents (Kutler, pp. 3, 5, 6, 8 and 193-194). Nixon wanted to break into the Brookings Institution, the Rand Corporation and the Council of Foreign Relations to steal national security information that he would release to politically tarnish the Democrats (Kutler, pp. 17, 24). John Erlichman wanted to break into the National Archives in order to steal secrets that they would use to tarnish their political enemies. Nixon wanted to recreate McCarthyism, with all its false accusations, dirty tricks and character assassinations, within the United States (Kutler, pp. 8, 11). In short, Nixon wanted to wage war against domestic political opposition. The plan involved using the IRS to harass political enemies. There was discussion of getting the secret service to spy on opposing candidates during the campaign. G. Gordon Liddy also got John Mitchell, Attorney General and head of the Committee to Re-elect the President, to sponsor a plan he drafted to disrupt, harass, and repress political enemies. Meanwhile, the economy was becoming wobbly as well. To compound that, balance of payments problems arose in the spring of 1971. By August, Nixon felt the United States had to leave the gold standard.

The post-break-in period can be considered in four phases. Phase One is depicted in Figure 1. In this phase, the Watergate burglars were arrested on June 17, 1972, on the anniversary of Nixon's re-endorsement of the Huston plan. Under the direction of G. Gordon Liddy and E. Howard Hunt, and supported with Republican Committee to Re-Elect the President funds, the Watergate Five -- McCord, Barker, Sturgis, Gonzales and Martinez -- were in the process of breaking and entering into the Democratic National Committee Headquarters when they were apprehended by Washington, D.C. police. Their immediate objective was to repair a bug they had installed in an earlier surreptitious entry. The real purpose was political espionage through a covert project that Liddy had codenamed GEMSTONE. Their bugs were planted in the Democratic National Committee offices of the Watergate Office Building. According to Frank Sturgis, one of the burglars who this writer interviewed in the 1970s, the objective was to find out whether Howard Hughes had given money to the Democratic party. They claimed their purposes were related to national security to cover their objectives. In fact, Nixon, Colson and Haldeman discussed ways of pinning it on the CIA. They tried to get the CIA to obstruct the FBI investigation of the matter (Kutler, pp. 61, 67-70). By the time these culprits were convicted, Nixon had concluded the Paris Peace Accords on January 27, 1973. Although this may have rendered Nixon's approval slightly more robust, the conviction of the Watergate burglars coincided with the beginning of the decline of Nixon's public approval.

In Phase Two, shown in Figure 2, Nixon's public approval continued to plummet as new information emerged implicating the White House and then Nixon himself in the cover-up. James McCord, surveillance-meister of the team, exposed a cover-up and perjury to Judge John Sirica. A little more than a year later, John Dean, Counsel to the President, revealed that Nixon had approved a request from Hunt and the burglars for hush money on June 23, 1972. Nixon knew where the money could be obtained. John Mitchell, then Attorney General, arranged to obtain the funds of silence from Thomas Pappas, in exchange for securing an Ambassadorship to Greece for Henry Tasca, a friend of his (Kutler, pp. 187, 218-219), while Frank Sturgis told acquaintances that it came from Robert Vesco. Moreover, Nixon explicitly advocated the cover story of telling the FBI that sensitive CIA assets were involved and that they should not push too hard on the investigation. Nixon was thus implicated in obstruction of justice. Afterward, Alexander Butterfield indicated that the events had been taped and any oral evidence from the White House Oval Office was on the tapes.

Phase Three was the struggle for evidence. Senator Sam Ervin, Chairman of the Senate Watergate Committee, and Archibald Cox, Special Prosecutor, sought the tapes. Nixon refused and the case went to the Supreme Court, which ruled that Nixon had to release the tapes. The October 1973 Oil Embargo and Production cutback increased the cost of production in the economy. On October 20th, Nixon ordered the firing of Special Prosecutor Archibald Cox. Attorney General Elliott Richardson and his assistant William Ruckelshaus refused to fire him, and were forced to resign in protest. Meanwhile, the plumbers team, which included most of the Watergate bugging team, was indicted for breaking into the office of Dr. Lewis Fielding, Daniel Ellsberg's psychiatrist. The exposure of this attempt to get enough dirt to destroy him for releasing the Pentagon Papers, which revealed how the U.S. became involved in the Vietnam War, is linked to Nixon's popularity sinking further.

In the final phase, the House deliberated impeachment and voted articles of impeachment. The articles charged Nixon with failure to take care that the laws were faithfully executed, and with abuse of power, obstruction of justice, and sabotage of the democratic process, in a manner that warranted his removal from office. With enough of the evidence disclosed, Nixon was warned that the Senate would convict him, lest he resign. Therefore, Nixon resigned the Presidency on August 8, 1974.

Figure 5 highlights the beginning of the fall of Richard Nixon. The criminalization of Nixon and his political demise is shown in a forecast profile. A forecast from this point performed with a Box-Jenkins time series analysis is also shown.

Throughout the scope of the Watergate scandal, the press was scrupulous about corroborating its leads and not printing false information. The prosecutors were scrupulous with regard to their conduct. There was no deluge of prosecutorial leaks. From an analysis of the post-impact change in the approval ratings, a model of the impact can be constructed.

A Box-Jenkins-Tiao ARIMA intervention analysis permits this researcher to model the impact on Nixon's fragile Gallup Poll presidential approval, from which it can be reduced to a nonlinear difference equation:

In simple English, apart from the regular autocorrelated approval, the approval rate is reduced by change in scandal by a factor of 12.36. The model shows that the influence of the scandal greatly depresses the approval rating. From the model developed, a forecast along with the upper and lower confidence intervals is projected forward into time and plotted. Figure 5 shows that the forecast cleaves tightly to the actual data once that has been gathered, that the model is good, and that it is thereby tested for predictive validity with satisfying results.

At this juncture, a caveat should be issued. Not all political crises follow the Watergate pattern. To develop a theory of political scandals, other scandals -- such as the hostage seizure during the Carter administration, the Iran-Contra scandal during the Reagan administration, and the current Lewinsky affair and impeachment trial -- would have to be examined. Patterns of presidential crisis approval ratings are found to differ. Nixon's ratings nose-dived after Hunt and Liddy were convicted, and were scraping the bottom when the House took up deliberation of impeachment. In contrast, President Clinton's presidential approval ratings have proved more robust and very resilient. After four months following the exposure of the Lewinsky affair, Clinton's Gallup Poll approval ratings began to trend upward. During the Senate impeachment trial, Clinton's presidential approval ratings were at 67 percent.

Many scholars -- including political scientist Edward Tufte -- have claimed that the state of the economy accounts for much public support of the president. During the Nixon era, the economy was shaky. There was the Oil Embargo and production cutback in October 1973, which increased the costs of production. During the Clinton era, the economy flourished, compared to those of other countries around the world. If the economy accounts for much political support, then Clinton could count on much more public support than Nixon could.

President Bill Clinton scored numerous triumphs in the foreign policy arena. He stopped a genocide in Bosnia (with the help of Ambassador Richard Holbrooke), he brokered two mid-East peace accords, facilitated an end to the Terrorist "Troubles" in Ireland, temporarily forestalled massacres in Kosovo, called for a global initiative against international and transnational terrorists at the UN, and offered aid to the Central American neighbors devastated by Hurricane Mitch. When Osama bin Laden atttacked US Embassies in Africa, Clinton identified the culprit and ordered swift retaliation. When Saddam violated the UN disarmament agreements and sought to develop weapons of mass destruction, President Clinton ordered military attacks upon the rogue regime of Iraq.

With the help of Congress, Clinton balanced the budget, waged war against the Tobacco Industry, and fought for campaign finance reform. He sought to bring the Internet into the schools and to hire more teachers. Whereas the people wanted Nixon to leave, the mass public wants Clinton to be neither impeached nor forced to resign. Notwithstanding Clinton's personal peccadilloes, 64 percent of the nation's adults, according to a January Gallup poll, do not want Clinton to be removed from office. Rather, 56 percent think he should be censured, and 54 percent of the public disapprove of how the Republicans are handling the investigation (Gallup Organization, 1999). This orientation is reflected in Clinton's high Gallup Poll presidential job approval over time, shown in Figure 6.

While Nixon's efforts in China and the Middle East were highlights of his foreign policy, the rapprochement with mainland China, while providing for leverage against Russia and North Vietnam, was seen as forsaking Taiwan, a longstanding ally. His handling of the Vietnam War was disastrous. Reviled by large segments of the population for its expansion into the neighboring country of Cambodia and his bombing campaigns against North Vietnam, his conclusion of the peace accords was welcomed by the American public. Unfortunately, the frail accords he agreed to in Paris dishonorably abandoned allies and left the regime in South Vietnam to catastrophic collapse upon invasion of the Viet Cong. In retrospect, architects of that war -- including former Secretary of Defense Robert McNamara and former National Security Advisor McGeorge Bundy -- have since confessed that the war was a terribly tragic mistake that should not be repeated.

It is not difficult to see why Clinton's job performance receives high approval by the voting public and why both impeachment and conviction would be contrary to the general will of the people.

References

Box, G.E.P., Jenkins, G. and Reinsel, G. (1994). Time Series Analysis: Forecasting and Control, 3rd ed., pp. 462-479.

Box, G.E.P. and Tiao (1975). "Intervention Analysis with Applications to Economic and Environmental Problems." Journal of the American Statistical Association, pp. 70, 70-92.

Kutler, S.I. (1997). Abuse of Power: The New Nixon Tapes. The Free Press, New York.

Ege, G. et al. (1993) SAS/ETS User's Guide, Version 6, 2nd ed., pp. 101-182.

Woodfield, T. (1987) "Time Series Intervention Analysis using SAS Software," SUGI Proceedings, pp. 331-339.

Woodward, D. (1997). SAS Institute, Inc. Cary, NC. November 18, 1997. Personal communication.

Dr. Yaffee is a senior statistician in ACF's Statistics and Social Sciences Group.
Date Posted: February 12, 1999
Keywords updated: August 9, 1999

Postcript

President William Jefferson Clinton was acquitted of all charges levied against him in the articles of impeachment the day this article was posted. His job performance approval may have predisposed many Senators to think that the seriousness of the charges against him did not rise to the level of removal from office (March, 1999).

This article is a thumbnail sketch of a portion of a chapter on intervention analysis found in the forthcoming book An Introduction to Time Series Analysis and Forecasting with Applications of SAS and SPSS (April 24, 2000) San Diego: Academic Press, by Robert A. Yaffee with Monnie McGee.

This article has been reviewed times since 7/30/99.