Economic Prosperity and Presidential Popularity: Sorting Out the Effects

• Author: David J. Lanoue
• DOI: 10.1177/106591298704000203
• Published date: 01 June 1987
• Date: 01 June 1987
• Subject Matter: Articles

ECONOMIC

PROSPERITY

AND

PRESIDENTIAL

POPULARITY:

SORTING

OUT

THE

EFFECTS

DAVID J.

LANOUE

University

of

Texas

at

El

Paso

GREAT

deal

of

effort

has

gone

into

determining

which

economic

vari-

ables

influence

presidential

popularity

and

the

public’s

voting

be-

~

JL

havior

with

respect

to

the

presidency.

The

major

variables

of

interest

through

the

years

have

been

inflation,

unemployment,

and

real

disposable

income.

From

these three

variables,

numerous

models

have been

tested,

with

results

ranging

from

finding

economic

effects

unimportant

(Stigler

1973)

to

finding

that

all

three

variables

have

significant

effects (Hibbs

1982).

Unfortunately,

less

work

has

gone

into

trying

to

determine

how

these

economic

variables

are

related

to

one

another,

and

how

these

relationships

affect

the

substantive

interpretations

of

our

models.

In

particular,

we

might

consider

the

relationship

between

unemployment

and

real

disposable

in-

come,

two

highly

correlated

variables.

Of

these

two

variables,

unemployment

has

received

the

greatest

atten-

tion

from

political

scientists.

This

is

not

surprising

when

one

considers

that

the

jobless

rate

is

one

of the

most

publicized

economic

statistics

in

the

na-

tional

news

media.

What

is

more

surprising

is

that

the

unemployment

rate

has

been

of

only

limited

value

as

a

predictor

of

the

popularity

of

the

presi-

dent

and

his

party.

Mueller

(1970,

1973)

found

a

significant

effect

on

popularity

using

a

variable

called &dquo;economic

slump&dquo;

(created

from

the

un-

employment

data).

His

results,

however,

have been

attacked

by

Hibbs

(1974),

who

controlled

for

residual

autocorrelation,

and

found

Mueller’s

&dquo;economic

slump&dquo;

variable

insignificant.

Rivers

(1980),

using

a

distributed

lag

model,

found

that

unemployment

had

an

effect

on

popularity

during

Dwight

Eisenhower’s

presidency.

MacKuen

(1983)

found

a

strong

immedi-

ate

effect

of

unemployment

on

popularity

between

1963

and

1980,

but

it

was

an

effect

that

wore

off

quickly.

Otherwise,

research

on

the

United

States

has

generally

found

no

significant

relationship

between

unemployment

and

presidential

popularity

(see,

for

example,

Cameron

1984;

Norpoth

1984;

and

Monroe

1978).

Research

involving

real

disposable

income

has

been

more

promising.

Kramer

(1971:

141)

found

it

to

be

&dquo;the

most

important&dquo;

variable

in

ex-

plaining

a

party’s

share

of

the

vote

in

congressional

elections.

Likewise,

Tufte

(1975)

discovered

that

the

change

in

income

levels

over

one

year

had

very

significant

effects

on

both

congressional

and

presidential

voting.

Bloom

Received:

January

29,

1986

Revision

Received:

June

4,

1986

Accepted

for

Publication:

June

5,

1986

NOTE:

The

author

would

like

to

thank

Subha

Ramachandran,

Peter

Schrott,

and

Jeffrey

Segal

for

their

helpful

comments

and

criticisms.

Special

thanks

are

due

to

Helmut

Nor-

poth

for

his

useful

suggestions,

as

well

as

the

use

of

his

data.