34 2: Th eories of Political Control of Bureaucracy
the anecdotal evidence to support such assertions is enormous, they are not
given much weight by the ten cases investigated for this study. (1994, 368)Th e clearest evidence of how individual bureaucrats respond to the problems
of divided government and co-management is in the splendid research of Marissa
Martino Golden (1992). Using a modifi ed version of Albert O. Hirschman’s con-
cepts (1970) of bureaucratic exit, voice, and loyalty, she studied bureaucrats in the
Civil Rights Division of the Department of Justice and in the National Highway
Transportation Safety Administration (NHTSA) of the Department of Transpor-
tation during the administration of a Republican president, Ronald Reagan, and
a Democratic Congress. It is important to remember that the Reagan administra-
tion was particularly antibureaucratic, advocated a strong policy departure from
the status quo, and was in a decided ideological disagreement with congressional
leaders as well as strong policy advocates in the bureaucracy.
Th e assumption that bureaucrats either cooperate or resist, Golden argues,
is too simplistic. She determined that promoting or inhibiting bureaucratic
resistance depended upon the bureaucrat’s careerist ideology, the dominant
agency profession (law, engineering), the agency’s esprit, the agency’s history,
and the confi dence of the careerists. In the Civil Rights Division, she found
a long-standing ideology of support for the rights of racial minorities and
women, a dominant profession (law), and a deep bureaucratic commitment
to enforcing the civil rights laws. In the face of the Reagan administration’s
attempts to dismantle the division or force it not to enforce these laws, Golden
found the widespread use of voice in several forms: a war of memos between
career attorneys and political appointees, some use of leaks and sabotage, and
very little cooperative action. Th ere was a good bit of exit, some of it so-called
creative resignation, to make a point. But exit was not taken lightly, and bu-
reaucrats carefully weighed the tradeoff between exit and the consequent loss
of voice. Th ere was loyalty, sometimes mixed with withdrawal or neglect. Oft en
voice and loyalty worked together, namely, the bureaucrat who stays on in part
because of the desire to be an infl uential policy voice.
At the NHTSA, the issue was air bags. Congress and the NHTSA wanted
them, but the automobile industry and the Reagan administration did not. A
rule calling for air bags was in place at the beginning of the Reagan administra-
tion. Th e rule was fi rst rescinded, then overruled by the Supreme Court, and then
taken “up” to the level of Transportation Secretary Elizabeth Dole and out of the
hands of NHTSA. Golden found some use of voice in attempts to infl uence pol-
icy, but much less than at the Civil Rights Division. Th ere were some leaks to
interest groups and Congress by career bureaucrats, but virtually no exit. Th ere
was a good bit of loyalty and steady policy implementation following presidential
leadership, as well as a good bit of passive behavior and neglect.
Overall, the bureaucrats in the Civil Rights Division, when compared with
those at NHTSA, were more ideological, shared a common esprit de corps, tended