Sabato's Crystal Ball

The Politics of White House Firings

Would giving Shinseki the axe be good for Obama's poll numbers?

Larry J. Sabato, Director, U.Va. Center for Politics May 29th, 2014

facebooktwittergoogle_plusredditpinterestmail
U.Va. Center for Politics Director Larry J. Sabato is contributing a regular column to Politico Magazine. In light of the recent scandal with the Veterans Affairs health system, this week he looks back at presidential firings to see if they actually proved beneficial to a president’s job approval rating.

In the piece, we listed 35 cases of significant changes in presidential administrations where press reports made clear the departures were not voluntary, stretching from the Truman administration to the present. We also asked readers to suggest additional examples we might have missed. There were two names that merited inclusion:

— In 1983, Secretary of the Interior James G. Watt resigned under duress after he made racially insensitive remarks. President Reagan’s Gallup approval rating went from 47% for the period of Sept. 16-19, 1983, to 49% for the period of Oct. 21-24, 1983, following Watt’s ouster on Oct. 9. Thanks to Twitter user @jaybennett1234 for suggesting this addition.

— More recently, Federal Emergency Management Agency Director Michael Brown resigned in 2005 after becoming the face of the Bush administration’s poor response to Hurricane Katrina. We largely restricted our initial list to Cabinet-level appointees and other major offices, but Brown belongs on the list, too. President Bush’s approval rating was at 46% in Gallup from Sept. 8-11, 2005, before Brown’s exit, and it fell to 40% in the Sept. 16-18 period. Thanks to Jocelyn G. Brown Hall for reminding us of “Heck of a job, Brownie.”

So to update our tallies of the impact of these presidential staffing shuffles:

— Of the 30 firings or sets of firings (some happened at the same time), in only five instances did the president’s poll numbers improve three or more points after the firing, and all of the gains appear to be minor and short-term.

— Thirteen firings actually saw the president’s job approval fall three or more points in the next Gallup poll, by an average of four percentage points.

— Twelve firings had no discernible impact on the president’s ratings. We considered a variation of one or two percentage points to be well within polling’s margin of error. It is also very possible other factors besides dismissals produced the variations in presidential approval, though in most cases, Gallup took a survey shortly after the headline event.

These additions give more credence to the overall findings in the piece: Personnel changes in the White House are not all that helpful in improving a president’s approval rating — and can even hurt by drawing more attention to the problem at hand.

The Editors

When presidential popularity sags, the predictable calls begin: Off with their heads! The heads, of course, belong to key administrators and Cabinet officials who are perceived to have contributed to whatever woes are besetting the president.

The calls come from both friends and foes of the White House. The friends want to satisfy and silence baying opponents with a pound of prominent flesh; the foes hope it’s just the first pound.

President Obama has been resistant to dismissing any senior person. He refused to fire Secretary of Health and Human Services Kathleen Sebelius during the disastrous Obamacare rollout (though Sebelius later resigned). Nor has he, as of this writing, dismissed Secretary of Veterans Affairs Eric Shinseki, despite widespread problems in the VA that are ballooning into a damaging scandal.

To read the rest of the column, please click here.