Description
The purpose of this paper is to establish the consensus about the tremendous economic
success of the Troubled Asset Relief Program (TARP) and explore theories of popular disapproval of
TARP.
Journal of Financial Economic Policy
Knowledge and the political economy of TARP
Michael E.S. Hoffman
Article information:
To cite this document:
Michael E.S. Hoffman, (2013),"Knowledge and the political economy of TARP", J ournal of Financial
Economic Policy, Vol. 5 Iss 3 pp. 300 - 312
Permanent link to this document:
http://dx.doi.org/10.1108/J FEP-10-2012-0047
Downloaded on: 24 January 2016, At: 21:47 (PT)
References: this document contains references to 24 other documents.
To copy this document: [email protected]
The fulltext of this document has been downloaded 99 times since 2013*
Users who downloaded this article also downloaded:
Wenling Lu, David A. Whidbee, (2013),"Bank structure and failure during the financial crisis", J ournal of
Financial Economic Policy, Vol. 5 Iss 3 pp. 281-299 http://dx.doi.org/10.1108/J FEP-02-2013-0006
Timo Tammi, (2013),"Are a culture of trust and morality associated with paying and repaying behavior?:
Lessons from developed countries 2004-2011", J ournal of Financial Economic Policy, Vol. 5 Iss 3 pp.
313-328 http://dx.doi.org/10.1108/J FEP-10-2012-0042
Eric Osei-Assibey, Baimba Augustine Bockarie, (2013),"Bank risks, capital and loan supply: evidence from
Sierra Leone", J ournal of Financial Economic Policy, Vol. 5 Iss 3 pp. 256-271 http://dx.doi.org/10.1108/
J FEP-09-2012-0041
Access to this document was granted through an Emerald subscription provided by emerald-srm:115632 []
For Authors
If you would like to write for this, or any other Emerald publication, then please use our Emerald for
Authors service information about how to choose which publication to write for and submission guidelines
are available for all. Please visit www.emeraldinsight.com/authors for more information.
About Emerald www.emeraldinsight.com
Emerald is a global publisher linking research and practice to the benefit of society. The company
manages a portfolio of more than 290 journals and over 2,350 books and book series volumes, as well as
providing an extensive range of online products and additional customer resources and services.
Emerald is both COUNTER 4 and TRANSFER compliant. The organization is a partner of the Committee
on Publication Ethics (COPE) and also works with Portico and the LOCKSS initiative for digital archive
preservation.
*Related content and download information correct at time of download.
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
7
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Knowledge and the political
economy of TARP
Michael E.S. Hoffman
Center for Economics, US Government Accountability Of?ce,
Washington, DC, USA
Abstract
Purpose – The purpose of this paper is to establish the consensus about the tremendous economic
success of the Troubled Asset Relief Program (TARP) and explore theories of popular disapproval of
TARP.
Design/methodology/approach – The analytical approach in this paper is a multivariate survey
logit based on two Pew Research Center surveys that include questions on knowledge of and views on
TARP. One survey is used to estimate a knowledge index of TARP that is applied to another survey to
estimate the impact of knowledge on opinions of TARP.
Findings – The author ?nds that knowledge of TARP is dependent in particular on education, party
af?liation, and sex. Controlling for partisan effects, views on TARP’s effectiveness are distorted by
limited knowledge of TARP in magnitudes that are politically signi?cant.
Practical implications – Despite the severity and dramatic spillovers associated with banking
crises, decisive interventions may prove dif?cult to defend in retrospect in light of ignorance and an
inability to conceptualize the nature of the counterfactual.
Originality/value – This paper contributes to our understanding of the political economy of an
important and controversial economic policy, in particular the roles of ideology and knowledge in
accounting for public opposition to TARP. TARP is largely misunderstood. The author estimates a
model of TARP knowledge based on this misconception, and shows that education, party
af?liation/leaning, and sex are important predictors of TARP knowledge. By applying this model of
TARP knowledge to a separate survey dataset, the author demonstrates that knowledge of TARP
(along with political ideology) is an important predictor of support for TARP. By integrating two
different surveys, he is able to better identify the role played by knowledge of TARP, rather than
simply demographic characteristics.
Keywords Government policy and regulation, Financial institutions and services, Financial economics,
Decision making, Microeconomics, Macroeconomic policy, Public ?nance, Macroeconomics,
Financial markets, Money and interest rates, Financial crises
Paper type Research paper
I. Introduction
The Troubled Asset Relief Program (TARP) is at once one of the most hated,
misunderstood, and effective policies in modern economic history. Explaining the
contrast between TARP’s performance and public and political opposition is a
challenge of political economy. In concert with other policies executed by the federal
government, TARP was responsible for restoring ?nancial stability at a time when
systemic failure in the banking systemthreatened to bring about a downturn of massive
scale and depth. Not only was ?nancial stability restored in short order (GAO, 2009),
The current issue and full text archive of this journal is available at
www.emeraldinsight.com/1757-6385.htm
JEL classi?cation – G28, D72, E63, E44, G01
The views here belong to the author and are not attributable to the US Government
Accountability Of?ce.
Journal of Financial Economic Policy
Vol. 5 No. 3, 2013
pp. 300-312
qEmerald Group Publishing Limited
1757-6385
DOI 10.1108/JFEP-10-2012-0047
JFEP
5,3
300
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
7
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
the ultimate budgetary cost is likely to be quite modest (CBO, 2011). I use two survey
datasets from the Pew Research Center to document the extent to which TARP is
misunderstood and believed to be ineffective. Furthermore, I investigate the root causes
of these opinions using extensive data on the demographics of survey respondents.
I ?nd that knowledge of TARP is dependent in particular on education, party af?liation,
and sex. I use these relationships to construct an index of TARP knowledge and apply it
to a separate dataset. Controlling for partisan effects, views on TARP’s effectiveness are
distorted by limited knowledge of TARP in magnitudes that are politically signi?cant.
The use of survey data to tease out more thorough explanations of policy views has
been common in the area of international trade (Scheve and Slaughter, 2001) and
subject to some work in broader economic policy (Blinder and Krueger, 2004). Several
other research efforts address aspects of the political economy of TARP. Congleton
(2009) provides an overview of the political economy of the ?nancial crisis and policy
response. Dorsch (2010) ?nds that support for TARP in Congress was driven by
campaign contributions and the importance of ?nancial services in legislators’
districts. Finally, Smith et al. (2011) argue that TARP can only be explained within a
context of “entangled political economy” in which the market and polity are
intertwined both before and after the crisis.
The rest of the paper is organized as follows. Section II summarizes key
assessments of TARP’s effectiveness and related public opinion. Section III identi?es a
number of hypotheses on why public opinion diverges from economic assessments.
Section IV describes the Pew surveys and details the analysis of survey response and
demographic data and estimation of TARP knowledge. Section V concludes.
II. Assessments of the TARP
There have been several outspoken and credible critics of the TARP program, most
notably Taylor (2010) and Stiglitz (2011), but their criticisms are narrow and do not
fundamentally challenge the conclusion that without TARP the economy would have
experienced a much more serious recession. The core of Taylor’s critique, outlined in a
simple event study in Taylor (2010), is that uncertainty about the federal response
(in particular uncertainty about the nature of TARP) led to large increases in instability
in the banking system in late September and early October 2008. However, Taylor
concedes that that the plan to purchase equity in ?nancial institutions over “toxic
assets” greatly improved ?nancial conditions (pp. 171-172). The ?rst program
introduced under TARP was a bank equity infusion program, initially rolled out for the
nine largest banks, known as the Capital Purchase Program (CPP). This program was a
deviation fromthe initial strategy to buy troubled mortgage-related assets directly from
?nancial institutions. It became clear that the latter approach would be too slow and
raised other implementation challenges, e.g. how much to pay for the assets. Equity
purchases, in contrast, were immediate and at transparent prices. The uncertainty that
plagued interbank markets in late September and early October resulted when
investors feared, at least temporarily, that the federal back stop for the banking system
was absent, ?rst when the House voted down an initial version of the TARP legislation
and later when investors feared that a program to purchase “toxic assets” would not be
executed in time. Stiglitz has leveled a variety of charges against TARP (several are
enumerated in Stiglitz, 2011) including that it lacked transparency, did not charge
banks enough for TARP funds, did not spur lending by recipient banks, and that it
Political
economy
of TARP
301
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
7
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
exacerbated moral hazard in the ?nancial system and solidi?ed the dominance of “too
big to fail” ?rms. Nevertheless, Stiglitz also concedes that CPP’s recapitalization of the
banks was necessary (p. 7).
Blinder and Zandi ?nd that the combined effects of the federal response (including
both government spending and ?nancial interventions) was to turn a potential
staggering loss of 7.9 percent of GDP to a 2.4 percent gain in 2009. Blinder and Zandi’s
evaluation of TARP and related interventions is calibrated by actual changes in
?nancial risk premia after the programs were announced. That these announcements
have been associated with considerable improvements in overall ?nancial stability
(GAO, 2009) and the health of the largest banks (Veronesi and Zingales, 2010) has been
established using event study methodology. Importantly, event studies and other
analyses are confounded by the fact that two other programs with similar goals, in
addition to CPP, were announced simultaneously: FDIC’s Temporary Liquidity
Guarantee Program (TLGP) to insure bank debt and the Federal Reserve’s Commercial
Paper Funding Facility (CPFF) to allow the fed to directly purchase commercial paper
from certain borrowers. While the simultaneity of these announcements prevents the
use of an event study methodology to isolate any one program’s effects, theory
suggests CPP had some unique features relative to the other two policies. The effects of
CPP are likely to be unique (and complementary) in that TLGP and CPFF address
liquidity concerns, while CPP focuses more directly on concerns about the solvency of
?nancial institutions – a key driver of banking crises (Gavin and Hausmann, 1998).
Supporting this view, the IMF (2009) found that capital injections were effective at the
height of the ?nancial crisis, but in some cases liquidity guarantees were not because
they “secure only a subset of creditors and not the bank as a whole.” To estimate the
collective impact of these related ?nancial interventions in the USA, Blinder and Zandi
simulate their model with only the ?scal response to the crisis. In this scenario GDP
still declines by 5 percent in 2009, implying that TARP and related interventions are
necessary to explain the bulk of the 2009 recovery[1].
Concerns voiced by Stiglitz, among others, about the impact of TARP and other
interventions on moral hazard are legitimate and serious. To the extent that TARP
strengthened expectations that future incidences of ?nancial distress would lead to
socialized rather than private losses (i.e. a “bailout”) creditors and shareholders
would have little incentive to limit risk taking. This view is insightful in regards to the
impact of moral hazard on future decision making, but not helpful in evaluating
the proximate impact of TARP on ?nancial stability and near term economic
growth. Furthermore, moral hazard can be addressed with a future “tax” or other
disincentive on future risk-taking, such as the increase in bank capital requirements for
systemically important ?nancial institutions envisioned under Basel III or the
Dodd-Frank Act.
The macroeconomic simulation and event study results stand in stark contrast to
public opinion. A June 2010 Pew Research Center survey found that a mere 38 percent
of respondents thought that TARP “helped prevent a more severe economic crisis”
while 54 percent did not. Similarly, according to a Bloomberg National Poll conducted
by Selzer & Company in October 2010 only 24 percent of respondents believed “TARP
[. . .] will eventually lead to a stronger economy” while 43 percent believed it would lead
to a weaker economy. More straightforward factual errors about TARP are also
widespread. In a July 2010 Pew Research Center survey, only a third of respondents
JFEP
5,3
302
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
7
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
correctly identi?ed George W. Bush as the president that signed TARP. In contrast,
54 percent identi?ed President Obama and 19 percent did not know the answer.
Americans often lack basic knowledge of political institutions and major political
Figures (Delli Carpini and Keeter, 1997).
III. Theories of TARP opinions
I have documented that TARP is both widely misunderstood and widely viewed as
ineffective. I suggest four interrelated (and perhaps interdependent) hypotheses
regarding the nature of popular opposition to TARP. Popular opinion may be
overwhelmingly against TARP because:
.
it is viewed as distributionally unfair (e.g. taxpayer dollars went to banks and
not those truly in need);
.
it is viewed through a partisan or ideological schema (e.g. party af?liation or
feelings about capitalism);
.
evaluations are based on incorrect information (e.g. that TARP was enacted by
President Obama); or
.
most people are unable to conceptualize the counterfactual (i.e. what would have
happened had TARP not been enacted).
The premise that TARP was distributionally unfair rests on two questionable
assertions:
(1) that TARP was expensive; and
(2) that TARP bene?ted only banks.
While the legislation enacting TARP authorized over $700 billion, total spending under
TARPis likely to be less than $450 billion, and the total budgetary cost is estimated to be
about $25 billion (CBO, 2011). As a result, something perceived to be a $700 billion
handout to banks is in fact a $25 billion program, where the net cost comes from
assistance to AIG (an insurance company), General Motors and Chrysler, and grants to
homeowners to reduce mortgage payments. Equity investments in banks have already
turned a pro?t (CBO, 2011). Perhaps more importantly, the distributional impact of
TARP is related to what would have happened in the absence of TARP. Research on
?nancial crises across countries (IMF, 2008) and a previous wave of bank failures in the
USA (Kupiec and Ramirez, 2010) indicates that the economic consequences of
bank-related downturns are felt broadly and deeply. These crises are more severe than
typical recessions (Reinhart and Rogoff, 2009), and the economic effects of de?ation and
unemployment on consumers are substantial. In contrast, $25 billion would have
amounted to a mere $223 per US household if distributed directly – and negligible
macroeconomic impact. Nevertheless, the perception that TARP was distributionally
unfair (regardless of the reality) is likely to have colored popular opinion of TARP’s
success.
TARP may be viewed unfavorably in hindsight because a considerable proportion
of individuals may automatically discount economic interventions – especially in
failing ?rms – as morally and economically inappropriate[2]. This schema may result
in relatively unfavorable assessments of TARP’s performance from republican or
libertarian-leaning individuals. I test this hypothesis in Section IV.
Political
economy
of TARP
303
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
7
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Individuals evaluate TARP based on the information they have. An appropriate
information set might include the time of enactment, economic conditions at enactment,
the economic outlook at enactment, future revisions to the economic outlook (more on
that below), and the precise details of the policy. If much of the information that
individuals have is incorrect, or incomplete, their ability to properly evaluate TARP
will be limited. In general, researchers have shown that ignorance can have large
effects on policy preferences and voting behavior, see especially Bartels (1996) and
Gilens (2001). I test this hypothesis with respect to TARP in Section IV.
Finally, TARP’s performance may be viewed negatively because individuals are
unable to conceptualize what might have happened without TARP (the counterfactual).
As noted above, historical precedents and macroeconomic simulations suggest that
absent sizeable government response to a systemic banking crisis, an even more severe
recession would have ensued. This is the counterfactual. Of course, the counterfactual is
not observed. Several important indicators of ?nancial stability were in principle
observable (if not observed) and drove economic commentary and media coverage at
the time. Most important among these were interbank spreads, or measures of
perceived creditworthiness within and among the banking system. Interbank spreads
(Figure 1) were a clear indication of the scale of ?nancial instability and, as the basis for
the event study by GAO (2009) and simulation results by Blinder and Zandi (2010),
were a clear indicator that ?nancial stability had been greatly improved. Studies of
retrospective voting establish that voters routinely use recent changes in economic
conditions to punish or reward incumbents (Fiorina, 1978; Wilkin et al., 1997; Campbell,
2005). While expert interpretation of interbank spreads could in principle have provided
voters with a readily available proxy for the counterfactual, in practice continuing
deterioration in other key economic indicators (notably unemployment) likely
contaminated the attempts of average individuals to conceptualize a without-TARP
counterfactual.
Figure 1.
Spread between three
month $US Libor and
three month treasury
(“TED spread”)
0
50
100
150
200
250
300
350
400
450
500
1
/
2
/
2
0
0
6
4
/
2
/
2
0
0
6
7
/
2
/
2
0
0
6
1
0
/
2
/
2
0
0
6
1
/
2
/
2
0
0
7
4
/
2
/
2
0
0
7
7
/
2
/
2
0
0
7
1
0
/
2
/
2
0
0
7
1
/
2
/
2
0
0
8
4
/
2
/
2
0
0
8
7
/
2
/
2
0
0
8
1
0
/
2
/
2
0
0
8
1
/
2
/
2
0
0
9
4
/
2
/
2
0
0
9
7
/
2
/
2
0
0
9
1
0
/
2
/
2
0
0
9
1
/
2
/
2
0
1
0
4
/
2
/
2
0
1
0
7
/
2
/
2
0
1
0
1
0
/
2
/
2
0
1
0
B
a
s
i
s
P
o
i
n
t
s
TED Spread
JFEP
5,3
304
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
7
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
IV. Survey data analysis
I analyze two Pew Research Center surveys, a June 2010 survey that included a
question about TARP effectiveness (“Thinking about the loans the federal government
provided to troubled banks and ?nancial institutions, which statement comes closer to
your view? The loans HELPED prevent a more severe economic crisis [OR] The loans
did NOT help prevent a more severe economic crisis.”), and a July 2010 survey that
included a question testing knowledge of TARP (“Was the government bailout of
banks and ?nancial institutions, also known as TARP, passed and signed into law
under [. . .] President George W. Bush [OR] President Barack Obama?”). The analysis
takes place in two steps. First, to better understand the roots of misinformation about
TARP, I estimate a model of TARP knowledge (the July 2010 question) based on the
demographic characteristics of survey respondents. Second, to estimate the role of
misinformation in unfavorable opinions about TARP, I apply the model of TARP
knowledge developed in the ?rst step to the question about TARP effectiveness in the
June 2010 survey. I hypothesize that better information leads to more favorable views
on TARP’s performance[3].
To develop a model of TARP knowledge, I transform the July 2010 survey question
into a dummy variable equal to 1 if respondents answered correctly (President
George W. Bush) or equal to zero if respondents answered President Barack Obama or
stated that they did not know the answer. The potential explanatory variables are age,
level of education, sex, party af?liation/leaning, race and employment status[4].
Summary statistics for both surveys are available in the Appendix. The regressions
are estimated using sample weights and with the population strati?ed by phone usage
(cell phone only, landline only, or both)[5]. As a robustness test, I also estimate a model
with an alternative measure of ?nancial knowledge, this one derived from two other
questions in the survey: one based on whether the respondent closely follows news
about ?nancial regulatory reform, and the other based on whether the respondent can
correctly identify the recipient of an EU bailout (Greece at the time). I call this measure
“?nancial awareness” and hypothesize that it captures knowledge of ?nancial markets
but does not directly measure knowledge of TARP. Table I contains the results of the
estimation of the “TARP knowledge” and “?nancial awareness” models, estimated
with the following regressors: level of education (less than high school, high school
graduate, some college, college graduate; less than high school is the reference
category), and party af?liation/leaning (republican or republican-leaning, democrat or
democrat-leaning, and independent (not leaning)/other; independent is the reference
category), sex (male is the reference category), and employment status (employed full
time is the reference category). Note that because ?nancial awareness is estimated as a
linear model (?nancial awareness is not dichotomous as is the TARP knowledge
measure), coef?cient values are not directly comparable across models.
The marginal effects indicate the practical signi?cance of these explanatory
variables. Based on the TARP knowledge model, college graduates are roughly
35 percent more likely to know that TARP was signed by President Bush than
individuals that did not graduate from high school. Individuals with a party af?liation
or that lean toward one party or the other are more informed than independents, with
republicans, for example, roughly 14 percent more likely to know that TARP was
signed by President Bush. Women are 13 percent less likely to know that TARP was
signed by President Bush (consistent with general ?ndings about political knowledge
Political
economy
of TARP
305
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
7
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
and women in Delli Carpini and Keeter (1997)). Results are broadly similar in the model
based on more general ?nancial awareness, indicating that these demographic
explanatory variables may be capturing factors that drive general knowledge of
economic policy and ?nancial markets.
In estimating the impact of TARP knowledge on opinions of TARP’s performance
I take the knowledge that TARP was signed by President Bush to be indicative of
broader and more accurate information about TARP. Of course, I do not directly
observe TARP knowledge in the June 2010 dataset. I use the model of TARP
knowledge I estimated above (based on July 2010 data) to approximate knowledge of
TARP in June 2010[6]. Because knowledge of TARP is estimated with error, this
method of estimating TARP knowledge will introduce an attenuation bias, which will
bias measured effects (coef?cients and test statistics) of TARP knowledge toward zero.
Thus, the practical and statistical signi?cance of TARP knowledge will be
underestimated by the model – results will represent a lower bound of the potential
impact of TARP knowledge on assessments of TARP’s effectiveness.
As a logit probability, TARP knowledge is an index from 0 to 100 re?ecting the
likelihood of knowing that President Bush signed TARP and presumably other
information relevant to assessing TARP (a potential information set is described in
Section III). I estimate a number of speci?cations, beginning with the variables
suggested by the theory outlined in Section III: TARP knowledge and party af?liation.
For each speci?cation, I estimate two variations, one for the model of TARP knowledge
and one for the model of ?nancial awareness. I also introduce two other factors
that may in?uence understanding of TARP’s effectiveness, education and a question
in the June 2010 survey about the extent to which the respondent follows news
Coef?cient (SE)
Explanatory
variable Dependent variable
TARP knowledge
(logistic model)
Financial awareness
(linear model)
Age 0.00795 (0.00659) 0.00274
* * *
(0.00080)
Education High school graduate 1.2496
* * *
(0.4504) 0.12612
* * *
(0.03714)
Some college 1.8352
* * *
(0.4452) 0.28073
* * *
(0.04064)
College graduate 1.9170
* * *
(0.4464) 0.43267
* * *
(0.03736)
Party af?liation Republican or republican-
leaning
0.5840
* *
(0.2899) 0.08789
* *
(0.03630)
Democrat or democrat-
leaning
0.5918
* *
(0.2877) 0.06043
*
(0.03371)
Sex Female 20.5531
* * *
(0.1779) 20.1264
* * *
(0.02280)
Employment
status
Employed part time 20.1384 (0.3188) 20.04324 (0.03672)
Retired 20.0689 (0.2716) 20.03144 (0.03225)
Not employed 0.3671 (0.2476) 20.01692 (0.03212)
Disabled 21.2586
*
(0.6601) 0.03130 (0.1024)
Student 20.2132 (0.9495) 0.007470 (0.1591)
Other employment status 210.8188
* * *
(1.0974) 20.05634 (0.04883)
Race Black 20.1545 (0.3145) 20.08386
* *
(0.03414)
Sample size 959 959
R
2
0.08 0.26
Note: Statistical signi?cant at:
*
10,
* *
5 and
* * *
1 percent levels, respectively
Table I.
Models of TARP
knowledge
JFEP
5,3
306
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
7
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
about the economy (“[Do you follow r]eports about the condition of the US
economy?” Closely/not closely). See results in Table II.
In models 1 and 4, party af?liation and both measures of ?nancial knowledge have a
statistically signi?cant impact on views of TARP’s effectiveness. The addition of
education (models 2 and 5) to the models add little explanatory power, and the behavior
of the standard errors and R
2
suggests some potentially unwelcome multicollinearity
with TARP knowledge[7]. The inclusion of news about the economy (models 3 and 6)
in the model also seems to add little. As noted above, news about the economy may be
dif?cult to interpret in the wake of TARP due to con?icting economic indicators.
Furthermore, Blinder and Krueger’s (2004) ?nding, that the source of news being
followed is as important as whether or not news is being followed at all, may also be
relevant here.
Based on model 1, the marginal effects of party and knowledge of TARP indicate
their practical importance. Democrats are roughly 29 percent more likely to believe that
TARP was helpful in averting a more severe economic crisis. The strength of this
partisan bias suggests the importance of controlling for party af?liation/leaning to
estimate the impact of TARP knowledge – republicans, for example, are more likely
than independents to know TARP was signed by President Bush (and this improves
their assessment of TARP), but improved knowledge is far outweighed by ideological
aversion to the policy. In addition, someone with a TARP knowledge index of 95 (i.e. a
95 percent chance of knowing that TARP was signed by President Bush) is 30 percent
more likely to think that TARP was effective than someone with a TARP knowledge
index of ?ve. Attenuation bias implies that the impact of TARP knowledge in this
scenario is at least 30 percent. In contrast, an increase in ?nancial awareness of similar
magnitude improves the assessment of TARP’s impact by only 24 percentage points.
Nevertheless, the model results using ?nancial awareness as the measure of knowledge
are qualitatively similar.
These estimates allow us to consider a number of counterfactuals. In the spirit of
McCloskey and Ziliak (1996), how politically signi?cant are these effects? In other
words, how would aggregate support for TARP appear in the absence of partisan bias,
the presence of perfect information, or both? Actual aggregate support for TARP is
about 41 percent[8]. Support under a scenario where all respondents had perfect
information (TARP knowledge index is 100) rises to 63 percent. Support under a
scenario where all respondents are independents (lacking either a clear ideological
aversion or attraction to TARP) leaves aggregate support roughly unchanged –
independents straddle the middle between republicans and democrats, who are in the
sample in similar proportion. Similarly, aggregate support under a scenario with no
partisan bias but perfect information is about 65 percent[9]. A sizable majority of
individuals support TARP with better information about the program. The magnitude
of the association between TARP knowledge and opinions of TARP’s performance is
quite signi?cant in political terms. Recall in addition that the attenuation bias suggests
that the magnitude of this effect is underestimated.
V. Conclusion
This paper contributes to our understanding of the political economy of an important
and controversial economic policy, in particular the roles of ideology and knowledge in
accounting for public opposition to TARP. Economic history has shown that policy
Political
economy
of TARP
307
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
7
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
C
o
e
f
?
c
i
e
n
t
(
S
E
)
T
A
R
P
k
n
o
w
l
e
d
g
e
F
i
n
a
n
c
i
a
l
a
w
a
r
e
n
e
s
s
E
x
p
l
a
n
a
t
o
r
y
v
a
r
i
a
b
l
e
M
o
d
e
l
1
M
o
d
e
l
2
M
o
d
e
l
3
M
o
d
e
l
4
M
o
d
e
l
5
M
o
d
e
l
6
T
A
R
P
k
n
o
w
l
e
d
g
e
1
.
4
3
4
3
*
*
(
0
.
7
0
4
0
)
1
.
5
1
9
0
(
0
.
9
7
2
7
)
1
.
4
1
8
2
*
*
(
0
.
7
0
2
6
)
F
i
n
a
n
c
i
a
l
a
w
a
r
e
n
e
s
s
1
.
1
3
2
1
*
*
(
0
.
5
2
2
9
)
0
.
7
3
1
8
(
1
.
0
9
3
7
)
1
.
1
2
7
7
*
*
(
0
.
5
2
4
4
)
E
c
o
n
o
m
y
n
e
w
s
0
.
0
8
0
4
(
0
.
2
1
0
7
)
0
.
0
1
4
5
(
0
.
2
0
9
2
)
E
d
u
c
a
t
i
o
n
H
i
g
h
s
c
h
o
o
l
g
r
a
d
u
a
t
e
2
0
.
4
7
6
9
(
0
.
4
0
4
4
)
2
0
.
2
4
6
1
(
0
.
3
8
3
6
)
S
o
m
e
c
o
l
l
e
g
e
2
0
.
5
1
8
7
(
0
.
4
6
6
6
)
2
0
.
2
2
0
9
(
0
.
4
5
9
4
)
C
o
l
l
e
g
e
g
r
a
d
u
a
t
e
2
0
.
0
7
5
1
(
0
.
4
7
8
3
)
0
.
0
5
6
9
(
0
.
6
0
0
5
)
P
a
r
t
y
a
f
?
l
i
a
t
i
o
n
D
e
m
o
c
r
a
t
o
r
d
e
m
o
c
r
a
t
-
l
e
a
n
i
n
g
1
.
1
9
9
5
*
*
*
(
0
.
1
8
6
0
)
1
.
2
0
7
1
*
*
*
(
0
.
1
8
9
0
)
1
.
2
0
6
3
*
*
*
(
0
.
1
8
7
5
)
1
.
2
2
6
4
*
*
*
(
0
.
1
8
7
8
)
1
.
2
0
1
7
*
*
*
(
0
.
1
9
6
6
)
1
.
2
2
7
5
*
*
*
(
0
.
1
8
8
9
)
I
n
d
e
p
e
n
d
e
n
t
/
o
t
h
e
r
0
.
7
2
3
8
*
*
(
0
.
3
0
3
4
)
0
.
7
2
4
9
*
*
(
0
.
3
1
3
7
)
0
.
7
4
0
4
*
*
(
0
.
3
0
6
8
)
0
.
6
5
9
*
*
(
0
.
2
8
8
9
)
0
.
5
9
5
8
*
(
0
.
3
0
7
4
)
0
.
6
6
1
9
*
*
(
0
.
2
9
1
8
)
S
a
m
p
l
e
s
i
z
e
8
6
0
8
6
0
8
6
0
8
7
7
8
7
7
8
7
7
R
2
0
.
0
7
0
.
0
8
0
.
0
7
0
.
0
7
0
.
0
8
0
.
0
7
N
o
t
e
:
S
t
a
t
i
s
t
i
c
a
l
s
i
g
n
i
?
c
a
n
t
a
t
:
*
1
0
,
*
*
5
a
n
d
*
*
*
1
p
e
r
c
e
n
t
l
e
v
e
l
s
,
r
e
s
p
e
c
t
i
v
e
l
y
Table II.
Logistic models of
TARP impact
JFEP
5,3
308
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
7
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
idleness in the face of a wave of bank failures results in severe economic dislocations.
Despite this compelling counterfactual, most people view TARP as a major policy
failure, notwithstanding its limited ?nancial cost. But while TARP is viewed as a
major policy failure, it is also largely misunderstood. Most people do not even know
that TARP was signed by President Bush, rather than his successor President Obama.
I estimate a model of TARP knowledge based this misconception, and show that
education, party af?liation/leaning, and sex are important predictors of TARP
knowledge. By applying this model of TARP knowledge to a separate survey dataset,
I demonstrate that knowledge of TARP (along with political ideology) is an important
predictor of support for TARP. By integrating two different surveys, I am able to better
identify the role played by knowledge of TARP, rather than simply demographic
characteristics. Individuals with better (predicted) knowledge of TARP may have
followed news about TARP more carefully or may have been better able to think
through the counterfactual – imagine the dif?culty of conducting an event study when
you are mistaken about when the event occurred. In the aggregate, I conduct a
counterfactual showing that improved knowledge of TARP across the electorate would
have been mirrored by broad popular support.
All this begs the question: how did TARP pass in the ?rst place? The policy has
been deeply unpopular and was electoral anathema within a few short months of its
passage. One simple – and I think, convincing – explanation is that members of
Congress fully understood the without-TARP counterfactual, even if their constituents
did not. And the without-TARP counterfactual was such a devastating recession that
its electoral consequences were potentially even more dire than supporting a policy
destined to be misunderstood and disliked. This conclusion provides a somewhat
reassuring assessment of the democratic process in the management of a ?nancial
crisis – the counterfactual can be persuasive to policy makers even when TARP’s
success remains largely unknown to voters.
Notes
1. The interaction between ?scal and ?nancial policy makes this explanation a bit of a
simpli?cation. In actuality, under ?nancial policy alone GDP declines by 4 percent, which is
more effective than ?scal policy alone, but much less than the combined effect of the policies
which yield the observed GDP growth of 2.4 percent in 2009. Thus, it is the interaction of the
two policies that generates the greatest economic bene?t.
2. As noted above, this view may be helpful for understanding the impact of moral hazard on
future decision making, but does not provide insight into the proximate impact of programs
like TARP on ?nancial stability and near term economic growth.
3. Pande (2011) surveys work on ?eld and natural experiments in voter information and argues
that better information helps voters hold politicians accountable.
4. I divide individuals into the following political categories: republicans or republican-leaning
independents, democrats or democratic-leaning independents, and independents who do not
lean toward either major party and those who claim other political af?liation.
5. The estimation procedures are available in SAS as proc surveylogistic and proc surveyreg.
Because sample weights are used to weight the covariance matrix, other adjustments – such
as for heteroskedasticity – are not feasible.
6. I consider the difference in available information in June vs July 2010 to be negligible.
Political
economy
of TARP
309
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
7
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
7. Education and party af?liation are of course both embedded in the index of TARPknowledge.
In theory, education may also capture the ability of individuals to process information,
not just their knowledge. As discussed above, party af?liation may capture ideological biases
in addition to differences in knowledge about politics or policy. It is important to note that
the introduction of all of the independent variables also embedded in the TARP knowledge
model would raise questions about model identi?cation, similar to those raised in a Heckman
correction model in the absence of a genuine instrument, in which case identi?cation rests
(weakly) of the functional form of the normal distribution (or in this case the logistic
distribution).
8. This proportion is slightly different from the one reported by Pew because I calculate it for
only those respondents where I have enough demographic information to run the model.
9. The slight difference between this scenario and the scenario with only perfect information is
due to the curvature of the logit function.
References
Bartels, L. (1996), “Uninformed votes: information effects in presidential elections”, American
Journal of Political Science, Vol. 40 No. 1, pp. 194-230.
Blinder, A.S. and Krueger, A. (2004), “What does the public know about economic policy and how
does it know it?”, Brookings Papers on Economic Activity, Vol. 2004 No. 1, pp. 327-387.
Blinder, A.S. and Zandi, M. (2010), How the Great Recession Was Brought to an End, Moody’s
Analytics, New York City, NY.
Campbell, J.E. (2005), “The fundamentals in US presidential elections: public opinion, the
economy and incumbency in the 2004 presidential election”, Journal of Elections, Public
Opinion, and Parties, Vol. 15 No. 1, pp. 73-83.
CBO (2011), Report on the Troubled Asset Relief Program, Congressional Budget Of?ce,
Washington, DC, October.
Congleton, R. (2009), “On the political economy of the ?nancial crisis and bailout of 2008-2009”,
Public Choice, Vol. 140 Nos 3/4, pp. 287-317.
Delli Carpini, M.X. and Keeter, S. (1997), What American Know About Politics and Why it
Matters, Yale University Press, New Haven, CT.
Dorsch, M. (2010), “Bailouts for sale? The political economy of the TARP bill”, unpublished
working paper.
Fiorina, M.P. (1978), “Economic and retrospective voting in American national elections”,
American Journal of Political Science, Vol. 22 No. 2, pp. 426-443.
GAO (2009), Troubled Asset Relief Program: One Year Later, Actions Are Needed to Address
Remaining Transparency and Accountability Challenges, Government Accountability
Of?ce, Washington, DC.
Gavin, M. and Hausmann, R. (1998), “The roots of banking crises: the macroeconomic context”,
IADB Working Paper No. 318, Inter-American Development Bank, Washington, DC.
Gilens, M. (2001), “Political ignorance and collective policy preferences”, The American Political
Science Review, Vol. 95 No. 2, pp. 379-396.
IMF (2008), “Financial stress and economic downturns”, World Economic Outlook, International
Monetary Fund, Washington, DC, October.
IMF (2009), “Market interventions during the ?nancial crisis: how effective and how to
disengage?”, Global Financial Stability Report, International Monetary Fund,
Washington, DC, October.
JFEP
5,3
310
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
7
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Kupiec, P.H. and Ramirez, C.D. (2010), “Bank failures and the cost of systemic risk: evidence from
1900-1930”, FDIC Center for Financial Research Working Paper, Federal Deposit
Insurance Corporation, Washington, DC.
McCloskey, D. and Ziliak, S. (1996), “The standard error of regressions”, Journal of Economic
Literature, Vol. 34 No. 1, pp. 97-114.
Pande, R. (2011), “Can informed voters enforce better governance? Experiments in low income
democracies”, Annual Review of Economics, Vol. 3, pp. 215-237.
Reinhart, C.M. and Rogoff, K. (2009), This Time is Different: Eight Centuries of Financial Folly,
Princeton University Press, Princeton, NJ.
Scheve, K.F. and Slaughter, M.J. (2001), “What determines individual trade-policy preferences?”,
Journal of International Economics, Vol. 54, pp. 267-292.
Smith, A., Wagner, R.E. and Yandle, B. (2011), “A theory of entangled political economy, with
application to TARP and NRA”, Public Choice, Vol. 148, pp. 45-66.
Stiglitz, J. (2011), Impact of TARP on Financial Stability, Testimony Before the Congressional
Oversight Panel, Washington, DC.
Taylor, J.B. (2010), “Getting back on track: macroeconomic policy lessons from the ?nancial
crisis”, Federal Reserve Bank of St Louis Review, Vol. 92 No. 3, pp. 165-176.
Veronesi, P. and Zingales, L. (2010), “Paulson’s gift”, Journal of Financial Economics, Vol. 97
No. 3, pp. 339-368.
Wilkin, S., Haller, B. and Norpoth, H. (1997), “From Argentina to Zambia: a worldwide test of
economic voting”, Electoral Studies, Vol. 16 No. 3, pp. 301-316.
Corresponding author
Michael E.S. Hoffman can be contacted at: [email protected]
(The Appendix follows overleaf.)
Political
economy
of TARP
311
To purchase reprints of this article please e-mail: [email protected]
Or visit our web site for further details: www.emeraldinsight.com/reprints
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
7
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Appendix
S
u
r
v
e
y
V
a
r
i
a
b
l
e
s
T
A
R
P
k
n
o
w
l
e
d
g
e
(
P
e
w
R
e
s
e
a
r
c
h
C
e
n
t
e
r
,
J
u
l
y
2
0
1
0
)
P
r
o
p
o
r
t
i
o
n
/
m
e
a
n
(
S
D
)
T
A
R
P
i
m
p
a
c
t
(
P
e
w
R
e
s
e
a
r
c
h
C
e
n
t
e
r
,
J
u
n
e
2
0
1
0
)
P
r
o
p
o
r
t
i
o
n
/
m
e
a
n
(
S
D
)
A
g
e
5
2
.
4
5
(
1
7
.
9
)
5
1
.
3
5
(
1
7
.
8
1
)
E
d
u
c
a
t
i
o
n
(
p
r
o
p
o
r
t
i
o
n
)
L
e
s
s
t
h
a
n
h
i
g
h
s
c
h
o
o
l
6
.
8
9
5
.
7
7
H
i
g
h
s
c
h
o
o
l
g
r
a
d
u
a
t
e
2
8
.
3
7
3
0
.
3
5
S
o
m
e
c
o
l
l
e
g
e
2
7
.
0
7
2
5
.
4
7
C
o
l
l
e
g
e
g
r
a
d
u
a
t
e
3
7
.
6
6
3
8
.
4
1
P
a
r
t
y
a
f
?
l
i
a
t
i
o
n
(
p
r
o
p
o
r
t
i
o
n
)
R
e
p
u
b
l
i
c
a
n
o
r
r
e
p
u
b
l
i
c
a
n
-
l
e
a
n
i
n
g
4
2
.
9
4
4
6
.
3
4
D
e
m
o
c
r
a
t
o
r
d
e
m
o
c
r
a
t
-
l
e
a
n
i
n
g
4
5
.
0
6
4
1
.
6
7
I
n
d
e
p
e
n
d
e
n
t
/
o
t
h
e
r
(
n
o
t
l
e
a
n
i
n
g
)
1
2
1
1
.
9
9
S
e
x
(
p
r
o
p
o
r
t
i
o
n
)
M
a
l
e
4
5
.
8
8
4
6
.
6
8
F
e
m
a
l
e
5
4
.
1
2
5
3
.
3
2
E
m
p
l
o
y
m
e
n
t
s
t
a
t
u
s
(
p
r
o
p
o
r
t
i
o
n
)
E
m
p
l
o
y
e
d
f
u
l
l
t
i
m
e
4
1
.
7
7
4
2
.
1
6
E
m
p
l
o
y
e
d
p
a
r
t
t
i
m
e
1
0
.
5
7
1
0
.
3
2
R
e
t
i
r
e
d
2
9
.
0
1
2
9
.
2
7
N
o
t
e
m
p
l
o
y
e
d
1
5
.
0
5
1
3
.
9
9
D
i
s
a
b
l
e
d
2
.
6
9
2
.
4
8
S
t
u
d
e
n
t
0
.
8
0
.
7
9
O
t
h
e
r
e
m
p
l
o
y
m
e
n
t
s
t
a
t
u
s
a
0
.
1
0
.
9
9
R
a
c
e
(
p
r
o
p
o
r
t
i
o
n
)
B
l
a
c
k
9
.
9
5
9
.
4
5
O
t
h
e
r
r
a
c
e
9
0
.
0
5
9
0
.
5
5
E
s
t
i
m
a
t
e
d
k
n
o
w
l
e
d
g
e
T
A
R
P
k
n
o
w
l
e
d
g
e
(
p
e
r
c
e
n
t
a
g
e
c
o
r
r
e
c
t
)
3
5
.
7
5
(
a
c
t
u
a
l
)
3
7
.
0
3
(
p
r
e
d
i
c
t
e
d
)
F
i
n
a
n
c
i
a
l
a
w
a
r
e
n
e
s
s
(
a
v
e
r
a
g
e
c
o
r
r
e
c
t
)
b
3
4
.
0
6
(
a
c
t
u
a
l
)
3
5
.
3
(
p
r
e
d
i
c
t
e
d
)
N
o
t
e
s
:
a
“
O
t
h
e
r
e
m
p
l
o
y
m
e
n
t
s
t
a
t
u
s
”
i
s
a
r
e
s
i
d
u
a
l
c
a
p
t
u
r
i
n
g
a
l
l
o
t
h
e
r
r
e
s
p
o
n
s
e
s
n
o
t
c
o
r
r
e
s
p
o
n
d
i
n
g
t
o
t
h
e
i
d
e
n
t
i
?
e
d
c
a
t
e
g
o
r
i
e
s
;
b
a
v
e
r
a
g
e
o
f
p
e
r
c
e
n
t
a
g
e
c
o
r
r
e
c
t
l
y
i
d
e
n
t
i
f
y
i
n
g
t
h
e
c
o
u
n
t
r
y
r
e
c
e
i
v
i
n
g
a
n
E
U
b
a
i
l
o
u
t
a
n
d
t
h
e
p
e
r
c
e
n
t
a
g
e
c
l
o
s
e
l
y
f
o
l
l
o
w
i
n
g
n
e
w
s
a
b
o
u
t
?
n
a
n
c
i
a
l
r
e
g
u
l
a
t
o
r
y
r
e
f
o
r
m
;
p
e
r
c
e
n
t
a
g
e
o
f
r
e
s
p
o
n
d
e
n
t
s
a
r
e
r
e
p
o
r
t
e
d
,
e
x
c
e
p
t
i
n
t
h
e
c
a
s
e
o
f
c
o
n
t
i
n
u
o
u
s
v
a
r
i
a
b
l
e
s
(
a
g
e
)
i
n
w
h
i
c
h
c
a
s
e
s
t
a
n
d
a
r
d
d
e
v
i
a
t
i
o
n
s
a
r
e
a
l
s
o
r
e
p
o
r
t
e
d
i
n
p
a
r
e
n
t
h
e
s
e
s
Table AI.
Summary statistics
JFEP
5,3
312
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
7
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
doc_917779233.pdf
The purpose of this paper is to establish the consensus about the tremendous economic
success of the Troubled Asset Relief Program (TARP) and explore theories of popular disapproval of
TARP.
Journal of Financial Economic Policy
Knowledge and the political economy of TARP
Michael E.S. Hoffman
Article information:
To cite this document:
Michael E.S. Hoffman, (2013),"Knowledge and the political economy of TARP", J ournal of Financial
Economic Policy, Vol. 5 Iss 3 pp. 300 - 312
Permanent link to this document:
http://dx.doi.org/10.1108/J FEP-10-2012-0047
Downloaded on: 24 January 2016, At: 21:47 (PT)
References: this document contains references to 24 other documents.
To copy this document: [email protected]
The fulltext of this document has been downloaded 99 times since 2013*
Users who downloaded this article also downloaded:
Wenling Lu, David A. Whidbee, (2013),"Bank structure and failure during the financial crisis", J ournal of
Financial Economic Policy, Vol. 5 Iss 3 pp. 281-299 http://dx.doi.org/10.1108/J FEP-02-2013-0006
Timo Tammi, (2013),"Are a culture of trust and morality associated with paying and repaying behavior?:
Lessons from developed countries 2004-2011", J ournal of Financial Economic Policy, Vol. 5 Iss 3 pp.
313-328 http://dx.doi.org/10.1108/J FEP-10-2012-0042
Eric Osei-Assibey, Baimba Augustine Bockarie, (2013),"Bank risks, capital and loan supply: evidence from
Sierra Leone", J ournal of Financial Economic Policy, Vol. 5 Iss 3 pp. 256-271 http://dx.doi.org/10.1108/
J FEP-09-2012-0041
Access to this document was granted through an Emerald subscription provided by emerald-srm:115632 []
For Authors
If you would like to write for this, or any other Emerald publication, then please use our Emerald for
Authors service information about how to choose which publication to write for and submission guidelines
are available for all. Please visit www.emeraldinsight.com/authors for more information.
About Emerald www.emeraldinsight.com
Emerald is a global publisher linking research and practice to the benefit of society. The company
manages a portfolio of more than 290 journals and over 2,350 books and book series volumes, as well as
providing an extensive range of online products and additional customer resources and services.
Emerald is both COUNTER 4 and TRANSFER compliant. The organization is a partner of the Committee
on Publication Ethics (COPE) and also works with Portico and the LOCKSS initiative for digital archive
preservation.
*Related content and download information correct at time of download.
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
7
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Knowledge and the political
economy of TARP
Michael E.S. Hoffman
Center for Economics, US Government Accountability Of?ce,
Washington, DC, USA
Abstract
Purpose – The purpose of this paper is to establish the consensus about the tremendous economic
success of the Troubled Asset Relief Program (TARP) and explore theories of popular disapproval of
TARP.
Design/methodology/approach – The analytical approach in this paper is a multivariate survey
logit based on two Pew Research Center surveys that include questions on knowledge of and views on
TARP. One survey is used to estimate a knowledge index of TARP that is applied to another survey to
estimate the impact of knowledge on opinions of TARP.
Findings – The author ?nds that knowledge of TARP is dependent in particular on education, party
af?liation, and sex. Controlling for partisan effects, views on TARP’s effectiveness are distorted by
limited knowledge of TARP in magnitudes that are politically signi?cant.
Practical implications – Despite the severity and dramatic spillovers associated with banking
crises, decisive interventions may prove dif?cult to defend in retrospect in light of ignorance and an
inability to conceptualize the nature of the counterfactual.
Originality/value – This paper contributes to our understanding of the political economy of an
important and controversial economic policy, in particular the roles of ideology and knowledge in
accounting for public opposition to TARP. TARP is largely misunderstood. The author estimates a
model of TARP knowledge based on this misconception, and shows that education, party
af?liation/leaning, and sex are important predictors of TARP knowledge. By applying this model of
TARP knowledge to a separate survey dataset, the author demonstrates that knowledge of TARP
(along with political ideology) is an important predictor of support for TARP. By integrating two
different surveys, he is able to better identify the role played by knowledge of TARP, rather than
simply demographic characteristics.
Keywords Government policy and regulation, Financial institutions and services, Financial economics,
Decision making, Microeconomics, Macroeconomic policy, Public ?nance, Macroeconomics,
Financial markets, Money and interest rates, Financial crises
Paper type Research paper
I. Introduction
The Troubled Asset Relief Program (TARP) is at once one of the most hated,
misunderstood, and effective policies in modern economic history. Explaining the
contrast between TARP’s performance and public and political opposition is a
challenge of political economy. In concert with other policies executed by the federal
government, TARP was responsible for restoring ?nancial stability at a time when
systemic failure in the banking systemthreatened to bring about a downturn of massive
scale and depth. Not only was ?nancial stability restored in short order (GAO, 2009),
The current issue and full text archive of this journal is available at
www.emeraldinsight.com/1757-6385.htm
JEL classi?cation – G28, D72, E63, E44, G01
The views here belong to the author and are not attributable to the US Government
Accountability Of?ce.
Journal of Financial Economic Policy
Vol. 5 No. 3, 2013
pp. 300-312
qEmerald Group Publishing Limited
1757-6385
DOI 10.1108/JFEP-10-2012-0047
JFEP
5,3
300
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
7
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
the ultimate budgetary cost is likely to be quite modest (CBO, 2011). I use two survey
datasets from the Pew Research Center to document the extent to which TARP is
misunderstood and believed to be ineffective. Furthermore, I investigate the root causes
of these opinions using extensive data on the demographics of survey respondents.
I ?nd that knowledge of TARP is dependent in particular on education, party af?liation,
and sex. I use these relationships to construct an index of TARP knowledge and apply it
to a separate dataset. Controlling for partisan effects, views on TARP’s effectiveness are
distorted by limited knowledge of TARP in magnitudes that are politically signi?cant.
The use of survey data to tease out more thorough explanations of policy views has
been common in the area of international trade (Scheve and Slaughter, 2001) and
subject to some work in broader economic policy (Blinder and Krueger, 2004). Several
other research efforts address aspects of the political economy of TARP. Congleton
(2009) provides an overview of the political economy of the ?nancial crisis and policy
response. Dorsch (2010) ?nds that support for TARP in Congress was driven by
campaign contributions and the importance of ?nancial services in legislators’
districts. Finally, Smith et al. (2011) argue that TARP can only be explained within a
context of “entangled political economy” in which the market and polity are
intertwined both before and after the crisis.
The rest of the paper is organized as follows. Section II summarizes key
assessments of TARP’s effectiveness and related public opinion. Section III identi?es a
number of hypotheses on why public opinion diverges from economic assessments.
Section IV describes the Pew surveys and details the analysis of survey response and
demographic data and estimation of TARP knowledge. Section V concludes.
II. Assessments of the TARP
There have been several outspoken and credible critics of the TARP program, most
notably Taylor (2010) and Stiglitz (2011), but their criticisms are narrow and do not
fundamentally challenge the conclusion that without TARP the economy would have
experienced a much more serious recession. The core of Taylor’s critique, outlined in a
simple event study in Taylor (2010), is that uncertainty about the federal response
(in particular uncertainty about the nature of TARP) led to large increases in instability
in the banking system in late September and early October 2008. However, Taylor
concedes that that the plan to purchase equity in ?nancial institutions over “toxic
assets” greatly improved ?nancial conditions (pp. 171-172). The ?rst program
introduced under TARP was a bank equity infusion program, initially rolled out for the
nine largest banks, known as the Capital Purchase Program (CPP). This program was a
deviation fromthe initial strategy to buy troubled mortgage-related assets directly from
?nancial institutions. It became clear that the latter approach would be too slow and
raised other implementation challenges, e.g. how much to pay for the assets. Equity
purchases, in contrast, were immediate and at transparent prices. The uncertainty that
plagued interbank markets in late September and early October resulted when
investors feared, at least temporarily, that the federal back stop for the banking system
was absent, ?rst when the House voted down an initial version of the TARP legislation
and later when investors feared that a program to purchase “toxic assets” would not be
executed in time. Stiglitz has leveled a variety of charges against TARP (several are
enumerated in Stiglitz, 2011) including that it lacked transparency, did not charge
banks enough for TARP funds, did not spur lending by recipient banks, and that it
Political
economy
of TARP
301
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
7
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
exacerbated moral hazard in the ?nancial system and solidi?ed the dominance of “too
big to fail” ?rms. Nevertheless, Stiglitz also concedes that CPP’s recapitalization of the
banks was necessary (p. 7).
Blinder and Zandi ?nd that the combined effects of the federal response (including
both government spending and ?nancial interventions) was to turn a potential
staggering loss of 7.9 percent of GDP to a 2.4 percent gain in 2009. Blinder and Zandi’s
evaluation of TARP and related interventions is calibrated by actual changes in
?nancial risk premia after the programs were announced. That these announcements
have been associated with considerable improvements in overall ?nancial stability
(GAO, 2009) and the health of the largest banks (Veronesi and Zingales, 2010) has been
established using event study methodology. Importantly, event studies and other
analyses are confounded by the fact that two other programs with similar goals, in
addition to CPP, were announced simultaneously: FDIC’s Temporary Liquidity
Guarantee Program (TLGP) to insure bank debt and the Federal Reserve’s Commercial
Paper Funding Facility (CPFF) to allow the fed to directly purchase commercial paper
from certain borrowers. While the simultaneity of these announcements prevents the
use of an event study methodology to isolate any one program’s effects, theory
suggests CPP had some unique features relative to the other two policies. The effects of
CPP are likely to be unique (and complementary) in that TLGP and CPFF address
liquidity concerns, while CPP focuses more directly on concerns about the solvency of
?nancial institutions – a key driver of banking crises (Gavin and Hausmann, 1998).
Supporting this view, the IMF (2009) found that capital injections were effective at the
height of the ?nancial crisis, but in some cases liquidity guarantees were not because
they “secure only a subset of creditors and not the bank as a whole.” To estimate the
collective impact of these related ?nancial interventions in the USA, Blinder and Zandi
simulate their model with only the ?scal response to the crisis. In this scenario GDP
still declines by 5 percent in 2009, implying that TARP and related interventions are
necessary to explain the bulk of the 2009 recovery[1].
Concerns voiced by Stiglitz, among others, about the impact of TARP and other
interventions on moral hazard are legitimate and serious. To the extent that TARP
strengthened expectations that future incidences of ?nancial distress would lead to
socialized rather than private losses (i.e. a “bailout”) creditors and shareholders
would have little incentive to limit risk taking. This view is insightful in regards to the
impact of moral hazard on future decision making, but not helpful in evaluating
the proximate impact of TARP on ?nancial stability and near term economic
growth. Furthermore, moral hazard can be addressed with a future “tax” or other
disincentive on future risk-taking, such as the increase in bank capital requirements for
systemically important ?nancial institutions envisioned under Basel III or the
Dodd-Frank Act.
The macroeconomic simulation and event study results stand in stark contrast to
public opinion. A June 2010 Pew Research Center survey found that a mere 38 percent
of respondents thought that TARP “helped prevent a more severe economic crisis”
while 54 percent did not. Similarly, according to a Bloomberg National Poll conducted
by Selzer & Company in October 2010 only 24 percent of respondents believed “TARP
[. . .] will eventually lead to a stronger economy” while 43 percent believed it would lead
to a weaker economy. More straightforward factual errors about TARP are also
widespread. In a July 2010 Pew Research Center survey, only a third of respondents
JFEP
5,3
302
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
7
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
correctly identi?ed George W. Bush as the president that signed TARP. In contrast,
54 percent identi?ed President Obama and 19 percent did not know the answer.
Americans often lack basic knowledge of political institutions and major political
Figures (Delli Carpini and Keeter, 1997).
III. Theories of TARP opinions
I have documented that TARP is both widely misunderstood and widely viewed as
ineffective. I suggest four interrelated (and perhaps interdependent) hypotheses
regarding the nature of popular opposition to TARP. Popular opinion may be
overwhelmingly against TARP because:
.
it is viewed as distributionally unfair (e.g. taxpayer dollars went to banks and
not those truly in need);
.
it is viewed through a partisan or ideological schema (e.g. party af?liation or
feelings about capitalism);
.
evaluations are based on incorrect information (e.g. that TARP was enacted by
President Obama); or
.
most people are unable to conceptualize the counterfactual (i.e. what would have
happened had TARP not been enacted).
The premise that TARP was distributionally unfair rests on two questionable
assertions:
(1) that TARP was expensive; and
(2) that TARP bene?ted only banks.
While the legislation enacting TARP authorized over $700 billion, total spending under
TARPis likely to be less than $450 billion, and the total budgetary cost is estimated to be
about $25 billion (CBO, 2011). As a result, something perceived to be a $700 billion
handout to banks is in fact a $25 billion program, where the net cost comes from
assistance to AIG (an insurance company), General Motors and Chrysler, and grants to
homeowners to reduce mortgage payments. Equity investments in banks have already
turned a pro?t (CBO, 2011). Perhaps more importantly, the distributional impact of
TARP is related to what would have happened in the absence of TARP. Research on
?nancial crises across countries (IMF, 2008) and a previous wave of bank failures in the
USA (Kupiec and Ramirez, 2010) indicates that the economic consequences of
bank-related downturns are felt broadly and deeply. These crises are more severe than
typical recessions (Reinhart and Rogoff, 2009), and the economic effects of de?ation and
unemployment on consumers are substantial. In contrast, $25 billion would have
amounted to a mere $223 per US household if distributed directly – and negligible
macroeconomic impact. Nevertheless, the perception that TARP was distributionally
unfair (regardless of the reality) is likely to have colored popular opinion of TARP’s
success.
TARP may be viewed unfavorably in hindsight because a considerable proportion
of individuals may automatically discount economic interventions – especially in
failing ?rms – as morally and economically inappropriate[2]. This schema may result
in relatively unfavorable assessments of TARP’s performance from republican or
libertarian-leaning individuals. I test this hypothesis in Section IV.
Political
economy
of TARP
303
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
7
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Individuals evaluate TARP based on the information they have. An appropriate
information set might include the time of enactment, economic conditions at enactment,
the economic outlook at enactment, future revisions to the economic outlook (more on
that below), and the precise details of the policy. If much of the information that
individuals have is incorrect, or incomplete, their ability to properly evaluate TARP
will be limited. In general, researchers have shown that ignorance can have large
effects on policy preferences and voting behavior, see especially Bartels (1996) and
Gilens (2001). I test this hypothesis with respect to TARP in Section IV.
Finally, TARP’s performance may be viewed negatively because individuals are
unable to conceptualize what might have happened without TARP (the counterfactual).
As noted above, historical precedents and macroeconomic simulations suggest that
absent sizeable government response to a systemic banking crisis, an even more severe
recession would have ensued. This is the counterfactual. Of course, the counterfactual is
not observed. Several important indicators of ?nancial stability were in principle
observable (if not observed) and drove economic commentary and media coverage at
the time. Most important among these were interbank spreads, or measures of
perceived creditworthiness within and among the banking system. Interbank spreads
(Figure 1) were a clear indication of the scale of ?nancial instability and, as the basis for
the event study by GAO (2009) and simulation results by Blinder and Zandi (2010),
were a clear indicator that ?nancial stability had been greatly improved. Studies of
retrospective voting establish that voters routinely use recent changes in economic
conditions to punish or reward incumbents (Fiorina, 1978; Wilkin et al., 1997; Campbell,
2005). While expert interpretation of interbank spreads could in principle have provided
voters with a readily available proxy for the counterfactual, in practice continuing
deterioration in other key economic indicators (notably unemployment) likely
contaminated the attempts of average individuals to conceptualize a without-TARP
counterfactual.
Figure 1.
Spread between three
month $US Libor and
three month treasury
(“TED spread”)
0
50
100
150
200
250
300
350
400
450
500
1
/
2
/
2
0
0
6
4
/
2
/
2
0
0
6
7
/
2
/
2
0
0
6
1
0
/
2
/
2
0
0
6
1
/
2
/
2
0
0
7
4
/
2
/
2
0
0
7
7
/
2
/
2
0
0
7
1
0
/
2
/
2
0
0
7
1
/
2
/
2
0
0
8
4
/
2
/
2
0
0
8
7
/
2
/
2
0
0
8
1
0
/
2
/
2
0
0
8
1
/
2
/
2
0
0
9
4
/
2
/
2
0
0
9
7
/
2
/
2
0
0
9
1
0
/
2
/
2
0
0
9
1
/
2
/
2
0
1
0
4
/
2
/
2
0
1
0
7
/
2
/
2
0
1
0
1
0
/
2
/
2
0
1
0
B
a
s
i
s
P
o
i
n
t
s
TED Spread
JFEP
5,3
304
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
7
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
IV. Survey data analysis
I analyze two Pew Research Center surveys, a June 2010 survey that included a
question about TARP effectiveness (“Thinking about the loans the federal government
provided to troubled banks and ?nancial institutions, which statement comes closer to
your view? The loans HELPED prevent a more severe economic crisis [OR] The loans
did NOT help prevent a more severe economic crisis.”), and a July 2010 survey that
included a question testing knowledge of TARP (“Was the government bailout of
banks and ?nancial institutions, also known as TARP, passed and signed into law
under [. . .] President George W. Bush [OR] President Barack Obama?”). The analysis
takes place in two steps. First, to better understand the roots of misinformation about
TARP, I estimate a model of TARP knowledge (the July 2010 question) based on the
demographic characteristics of survey respondents. Second, to estimate the role of
misinformation in unfavorable opinions about TARP, I apply the model of TARP
knowledge developed in the ?rst step to the question about TARP effectiveness in the
June 2010 survey. I hypothesize that better information leads to more favorable views
on TARP’s performance[3].
To develop a model of TARP knowledge, I transform the July 2010 survey question
into a dummy variable equal to 1 if respondents answered correctly (President
George W. Bush) or equal to zero if respondents answered President Barack Obama or
stated that they did not know the answer. The potential explanatory variables are age,
level of education, sex, party af?liation/leaning, race and employment status[4].
Summary statistics for both surveys are available in the Appendix. The regressions
are estimated using sample weights and with the population strati?ed by phone usage
(cell phone only, landline only, or both)[5]. As a robustness test, I also estimate a model
with an alternative measure of ?nancial knowledge, this one derived from two other
questions in the survey: one based on whether the respondent closely follows news
about ?nancial regulatory reform, and the other based on whether the respondent can
correctly identify the recipient of an EU bailout (Greece at the time). I call this measure
“?nancial awareness” and hypothesize that it captures knowledge of ?nancial markets
but does not directly measure knowledge of TARP. Table I contains the results of the
estimation of the “TARP knowledge” and “?nancial awareness” models, estimated
with the following regressors: level of education (less than high school, high school
graduate, some college, college graduate; less than high school is the reference
category), and party af?liation/leaning (republican or republican-leaning, democrat or
democrat-leaning, and independent (not leaning)/other; independent is the reference
category), sex (male is the reference category), and employment status (employed full
time is the reference category). Note that because ?nancial awareness is estimated as a
linear model (?nancial awareness is not dichotomous as is the TARP knowledge
measure), coef?cient values are not directly comparable across models.
The marginal effects indicate the practical signi?cance of these explanatory
variables. Based on the TARP knowledge model, college graduates are roughly
35 percent more likely to know that TARP was signed by President Bush than
individuals that did not graduate from high school. Individuals with a party af?liation
or that lean toward one party or the other are more informed than independents, with
republicans, for example, roughly 14 percent more likely to know that TARP was
signed by President Bush. Women are 13 percent less likely to know that TARP was
signed by President Bush (consistent with general ?ndings about political knowledge
Political
economy
of TARP
305
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
7
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
and women in Delli Carpini and Keeter (1997)). Results are broadly similar in the model
based on more general ?nancial awareness, indicating that these demographic
explanatory variables may be capturing factors that drive general knowledge of
economic policy and ?nancial markets.
In estimating the impact of TARP knowledge on opinions of TARP’s performance
I take the knowledge that TARP was signed by President Bush to be indicative of
broader and more accurate information about TARP. Of course, I do not directly
observe TARP knowledge in the June 2010 dataset. I use the model of TARP
knowledge I estimated above (based on July 2010 data) to approximate knowledge of
TARP in June 2010[6]. Because knowledge of TARP is estimated with error, this
method of estimating TARP knowledge will introduce an attenuation bias, which will
bias measured effects (coef?cients and test statistics) of TARP knowledge toward zero.
Thus, the practical and statistical signi?cance of TARP knowledge will be
underestimated by the model – results will represent a lower bound of the potential
impact of TARP knowledge on assessments of TARP’s effectiveness.
As a logit probability, TARP knowledge is an index from 0 to 100 re?ecting the
likelihood of knowing that President Bush signed TARP and presumably other
information relevant to assessing TARP (a potential information set is described in
Section III). I estimate a number of speci?cations, beginning with the variables
suggested by the theory outlined in Section III: TARP knowledge and party af?liation.
For each speci?cation, I estimate two variations, one for the model of TARP knowledge
and one for the model of ?nancial awareness. I also introduce two other factors
that may in?uence understanding of TARP’s effectiveness, education and a question
in the June 2010 survey about the extent to which the respondent follows news
Coef?cient (SE)
Explanatory
variable Dependent variable
TARP knowledge
(logistic model)
Financial awareness
(linear model)
Age 0.00795 (0.00659) 0.00274
* * *
(0.00080)
Education High school graduate 1.2496
* * *
(0.4504) 0.12612
* * *
(0.03714)
Some college 1.8352
* * *
(0.4452) 0.28073
* * *
(0.04064)
College graduate 1.9170
* * *
(0.4464) 0.43267
* * *
(0.03736)
Party af?liation Republican or republican-
leaning
0.5840
* *
(0.2899) 0.08789
* *
(0.03630)
Democrat or democrat-
leaning
0.5918
* *
(0.2877) 0.06043
*
(0.03371)
Sex Female 20.5531
* * *
(0.1779) 20.1264
* * *
(0.02280)
Employment
status
Employed part time 20.1384 (0.3188) 20.04324 (0.03672)
Retired 20.0689 (0.2716) 20.03144 (0.03225)
Not employed 0.3671 (0.2476) 20.01692 (0.03212)
Disabled 21.2586
*
(0.6601) 0.03130 (0.1024)
Student 20.2132 (0.9495) 0.007470 (0.1591)
Other employment status 210.8188
* * *
(1.0974) 20.05634 (0.04883)
Race Black 20.1545 (0.3145) 20.08386
* *
(0.03414)
Sample size 959 959
R
2
0.08 0.26
Note: Statistical signi?cant at:
*
10,
* *
5 and
* * *
1 percent levels, respectively
Table I.
Models of TARP
knowledge
JFEP
5,3
306
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
7
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
about the economy (“[Do you follow r]eports about the condition of the US
economy?” Closely/not closely). See results in Table II.
In models 1 and 4, party af?liation and both measures of ?nancial knowledge have a
statistically signi?cant impact on views of TARP’s effectiveness. The addition of
education (models 2 and 5) to the models add little explanatory power, and the behavior
of the standard errors and R
2
suggests some potentially unwelcome multicollinearity
with TARP knowledge[7]. The inclusion of news about the economy (models 3 and 6)
in the model also seems to add little. As noted above, news about the economy may be
dif?cult to interpret in the wake of TARP due to con?icting economic indicators.
Furthermore, Blinder and Krueger’s (2004) ?nding, that the source of news being
followed is as important as whether or not news is being followed at all, may also be
relevant here.
Based on model 1, the marginal effects of party and knowledge of TARP indicate
their practical importance. Democrats are roughly 29 percent more likely to believe that
TARP was helpful in averting a more severe economic crisis. The strength of this
partisan bias suggests the importance of controlling for party af?liation/leaning to
estimate the impact of TARP knowledge – republicans, for example, are more likely
than independents to know TARP was signed by President Bush (and this improves
their assessment of TARP), but improved knowledge is far outweighed by ideological
aversion to the policy. In addition, someone with a TARP knowledge index of 95 (i.e. a
95 percent chance of knowing that TARP was signed by President Bush) is 30 percent
more likely to think that TARP was effective than someone with a TARP knowledge
index of ?ve. Attenuation bias implies that the impact of TARP knowledge in this
scenario is at least 30 percent. In contrast, an increase in ?nancial awareness of similar
magnitude improves the assessment of TARP’s impact by only 24 percentage points.
Nevertheless, the model results using ?nancial awareness as the measure of knowledge
are qualitatively similar.
These estimates allow us to consider a number of counterfactuals. In the spirit of
McCloskey and Ziliak (1996), how politically signi?cant are these effects? In other
words, how would aggregate support for TARP appear in the absence of partisan bias,
the presence of perfect information, or both? Actual aggregate support for TARP is
about 41 percent[8]. Support under a scenario where all respondents had perfect
information (TARP knowledge index is 100) rises to 63 percent. Support under a
scenario where all respondents are independents (lacking either a clear ideological
aversion or attraction to TARP) leaves aggregate support roughly unchanged –
independents straddle the middle between republicans and democrats, who are in the
sample in similar proportion. Similarly, aggregate support under a scenario with no
partisan bias but perfect information is about 65 percent[9]. A sizable majority of
individuals support TARP with better information about the program. The magnitude
of the association between TARP knowledge and opinions of TARP’s performance is
quite signi?cant in political terms. Recall in addition that the attenuation bias suggests
that the magnitude of this effect is underestimated.
V. Conclusion
This paper contributes to our understanding of the political economy of an important
and controversial economic policy, in particular the roles of ideology and knowledge in
accounting for public opposition to TARP. Economic history has shown that policy
Political
economy
of TARP
307
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
7
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
C
o
e
f
?
c
i
e
n
t
(
S
E
)
T
A
R
P
k
n
o
w
l
e
d
g
e
F
i
n
a
n
c
i
a
l
a
w
a
r
e
n
e
s
s
E
x
p
l
a
n
a
t
o
r
y
v
a
r
i
a
b
l
e
M
o
d
e
l
1
M
o
d
e
l
2
M
o
d
e
l
3
M
o
d
e
l
4
M
o
d
e
l
5
M
o
d
e
l
6
T
A
R
P
k
n
o
w
l
e
d
g
e
1
.
4
3
4
3
*
*
(
0
.
7
0
4
0
)
1
.
5
1
9
0
(
0
.
9
7
2
7
)
1
.
4
1
8
2
*
*
(
0
.
7
0
2
6
)
F
i
n
a
n
c
i
a
l
a
w
a
r
e
n
e
s
s
1
.
1
3
2
1
*
*
(
0
.
5
2
2
9
)
0
.
7
3
1
8
(
1
.
0
9
3
7
)
1
.
1
2
7
7
*
*
(
0
.
5
2
4
4
)
E
c
o
n
o
m
y
n
e
w
s
0
.
0
8
0
4
(
0
.
2
1
0
7
)
0
.
0
1
4
5
(
0
.
2
0
9
2
)
E
d
u
c
a
t
i
o
n
H
i
g
h
s
c
h
o
o
l
g
r
a
d
u
a
t
e
2
0
.
4
7
6
9
(
0
.
4
0
4
4
)
2
0
.
2
4
6
1
(
0
.
3
8
3
6
)
S
o
m
e
c
o
l
l
e
g
e
2
0
.
5
1
8
7
(
0
.
4
6
6
6
)
2
0
.
2
2
0
9
(
0
.
4
5
9
4
)
C
o
l
l
e
g
e
g
r
a
d
u
a
t
e
2
0
.
0
7
5
1
(
0
.
4
7
8
3
)
0
.
0
5
6
9
(
0
.
6
0
0
5
)
P
a
r
t
y
a
f
?
l
i
a
t
i
o
n
D
e
m
o
c
r
a
t
o
r
d
e
m
o
c
r
a
t
-
l
e
a
n
i
n
g
1
.
1
9
9
5
*
*
*
(
0
.
1
8
6
0
)
1
.
2
0
7
1
*
*
*
(
0
.
1
8
9
0
)
1
.
2
0
6
3
*
*
*
(
0
.
1
8
7
5
)
1
.
2
2
6
4
*
*
*
(
0
.
1
8
7
8
)
1
.
2
0
1
7
*
*
*
(
0
.
1
9
6
6
)
1
.
2
2
7
5
*
*
*
(
0
.
1
8
8
9
)
I
n
d
e
p
e
n
d
e
n
t
/
o
t
h
e
r
0
.
7
2
3
8
*
*
(
0
.
3
0
3
4
)
0
.
7
2
4
9
*
*
(
0
.
3
1
3
7
)
0
.
7
4
0
4
*
*
(
0
.
3
0
6
8
)
0
.
6
5
9
*
*
(
0
.
2
8
8
9
)
0
.
5
9
5
8
*
(
0
.
3
0
7
4
)
0
.
6
6
1
9
*
*
(
0
.
2
9
1
8
)
S
a
m
p
l
e
s
i
z
e
8
6
0
8
6
0
8
6
0
8
7
7
8
7
7
8
7
7
R
2
0
.
0
7
0
.
0
8
0
.
0
7
0
.
0
7
0
.
0
8
0
.
0
7
N
o
t
e
:
S
t
a
t
i
s
t
i
c
a
l
s
i
g
n
i
?
c
a
n
t
a
t
:
*
1
0
,
*
*
5
a
n
d
*
*
*
1
p
e
r
c
e
n
t
l
e
v
e
l
s
,
r
e
s
p
e
c
t
i
v
e
l
y
Table II.
Logistic models of
TARP impact
JFEP
5,3
308
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
7
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
idleness in the face of a wave of bank failures results in severe economic dislocations.
Despite this compelling counterfactual, most people view TARP as a major policy
failure, notwithstanding its limited ?nancial cost. But while TARP is viewed as a
major policy failure, it is also largely misunderstood. Most people do not even know
that TARP was signed by President Bush, rather than his successor President Obama.
I estimate a model of TARP knowledge based this misconception, and show that
education, party af?liation/leaning, and sex are important predictors of TARP
knowledge. By applying this model of TARP knowledge to a separate survey dataset,
I demonstrate that knowledge of TARP (along with political ideology) is an important
predictor of support for TARP. By integrating two different surveys, I am able to better
identify the role played by knowledge of TARP, rather than simply demographic
characteristics. Individuals with better (predicted) knowledge of TARP may have
followed news about TARP more carefully or may have been better able to think
through the counterfactual – imagine the dif?culty of conducting an event study when
you are mistaken about when the event occurred. In the aggregate, I conduct a
counterfactual showing that improved knowledge of TARP across the electorate would
have been mirrored by broad popular support.
All this begs the question: how did TARP pass in the ?rst place? The policy has
been deeply unpopular and was electoral anathema within a few short months of its
passage. One simple – and I think, convincing – explanation is that members of
Congress fully understood the without-TARP counterfactual, even if their constituents
did not. And the without-TARP counterfactual was such a devastating recession that
its electoral consequences were potentially even more dire than supporting a policy
destined to be misunderstood and disliked. This conclusion provides a somewhat
reassuring assessment of the democratic process in the management of a ?nancial
crisis – the counterfactual can be persuasive to policy makers even when TARP’s
success remains largely unknown to voters.
Notes
1. The interaction between ?scal and ?nancial policy makes this explanation a bit of a
simpli?cation. In actuality, under ?nancial policy alone GDP declines by 4 percent, which is
more effective than ?scal policy alone, but much less than the combined effect of the policies
which yield the observed GDP growth of 2.4 percent in 2009. Thus, it is the interaction of the
two policies that generates the greatest economic bene?t.
2. As noted above, this view may be helpful for understanding the impact of moral hazard on
future decision making, but does not provide insight into the proximate impact of programs
like TARP on ?nancial stability and near term economic growth.
3. Pande (2011) surveys work on ?eld and natural experiments in voter information and argues
that better information helps voters hold politicians accountable.
4. I divide individuals into the following political categories: republicans or republican-leaning
independents, democrats or democratic-leaning independents, and independents who do not
lean toward either major party and those who claim other political af?liation.
5. The estimation procedures are available in SAS as proc surveylogistic and proc surveyreg.
Because sample weights are used to weight the covariance matrix, other adjustments – such
as for heteroskedasticity – are not feasible.
6. I consider the difference in available information in June vs July 2010 to be negligible.
Political
economy
of TARP
309
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
7
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
7. Education and party af?liation are of course both embedded in the index of TARPknowledge.
In theory, education may also capture the ability of individuals to process information,
not just their knowledge. As discussed above, party af?liation may capture ideological biases
in addition to differences in knowledge about politics or policy. It is important to note that
the introduction of all of the independent variables also embedded in the TARP knowledge
model would raise questions about model identi?cation, similar to those raised in a Heckman
correction model in the absence of a genuine instrument, in which case identi?cation rests
(weakly) of the functional form of the normal distribution (or in this case the logistic
distribution).
8. This proportion is slightly different from the one reported by Pew because I calculate it for
only those respondents where I have enough demographic information to run the model.
9. The slight difference between this scenario and the scenario with only perfect information is
due to the curvature of the logit function.
References
Bartels, L. (1996), “Uninformed votes: information effects in presidential elections”, American
Journal of Political Science, Vol. 40 No. 1, pp. 194-230.
Blinder, A.S. and Krueger, A. (2004), “What does the public know about economic policy and how
does it know it?”, Brookings Papers on Economic Activity, Vol. 2004 No. 1, pp. 327-387.
Blinder, A.S. and Zandi, M. (2010), How the Great Recession Was Brought to an End, Moody’s
Analytics, New York City, NY.
Campbell, J.E. (2005), “The fundamentals in US presidential elections: public opinion, the
economy and incumbency in the 2004 presidential election”, Journal of Elections, Public
Opinion, and Parties, Vol. 15 No. 1, pp. 73-83.
CBO (2011), Report on the Troubled Asset Relief Program, Congressional Budget Of?ce,
Washington, DC, October.
Congleton, R. (2009), “On the political economy of the ?nancial crisis and bailout of 2008-2009”,
Public Choice, Vol. 140 Nos 3/4, pp. 287-317.
Delli Carpini, M.X. and Keeter, S. (1997), What American Know About Politics and Why it
Matters, Yale University Press, New Haven, CT.
Dorsch, M. (2010), “Bailouts for sale? The political economy of the TARP bill”, unpublished
working paper.
Fiorina, M.P. (1978), “Economic and retrospective voting in American national elections”,
American Journal of Political Science, Vol. 22 No. 2, pp. 426-443.
GAO (2009), Troubled Asset Relief Program: One Year Later, Actions Are Needed to Address
Remaining Transparency and Accountability Challenges, Government Accountability
Of?ce, Washington, DC.
Gavin, M. and Hausmann, R. (1998), “The roots of banking crises: the macroeconomic context”,
IADB Working Paper No. 318, Inter-American Development Bank, Washington, DC.
Gilens, M. (2001), “Political ignorance and collective policy preferences”, The American Political
Science Review, Vol. 95 No. 2, pp. 379-396.
IMF (2008), “Financial stress and economic downturns”, World Economic Outlook, International
Monetary Fund, Washington, DC, October.
IMF (2009), “Market interventions during the ?nancial crisis: how effective and how to
disengage?”, Global Financial Stability Report, International Monetary Fund,
Washington, DC, October.
JFEP
5,3
310
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
7
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Kupiec, P.H. and Ramirez, C.D. (2010), “Bank failures and the cost of systemic risk: evidence from
1900-1930”, FDIC Center for Financial Research Working Paper, Federal Deposit
Insurance Corporation, Washington, DC.
McCloskey, D. and Ziliak, S. (1996), “The standard error of regressions”, Journal of Economic
Literature, Vol. 34 No. 1, pp. 97-114.
Pande, R. (2011), “Can informed voters enforce better governance? Experiments in low income
democracies”, Annual Review of Economics, Vol. 3, pp. 215-237.
Reinhart, C.M. and Rogoff, K. (2009), This Time is Different: Eight Centuries of Financial Folly,
Princeton University Press, Princeton, NJ.
Scheve, K.F. and Slaughter, M.J. (2001), “What determines individual trade-policy preferences?”,
Journal of International Economics, Vol. 54, pp. 267-292.
Smith, A., Wagner, R.E. and Yandle, B. (2011), “A theory of entangled political economy, with
application to TARP and NRA”, Public Choice, Vol. 148, pp. 45-66.
Stiglitz, J. (2011), Impact of TARP on Financial Stability, Testimony Before the Congressional
Oversight Panel, Washington, DC.
Taylor, J.B. (2010), “Getting back on track: macroeconomic policy lessons from the ?nancial
crisis”, Federal Reserve Bank of St Louis Review, Vol. 92 No. 3, pp. 165-176.
Veronesi, P. and Zingales, L. (2010), “Paulson’s gift”, Journal of Financial Economics, Vol. 97
No. 3, pp. 339-368.
Wilkin, S., Haller, B. and Norpoth, H. (1997), “From Argentina to Zambia: a worldwide test of
economic voting”, Electoral Studies, Vol. 16 No. 3, pp. 301-316.
Corresponding author
Michael E.S. Hoffman can be contacted at: [email protected]
(The Appendix follows overleaf.)
Political
economy
of TARP
311
To purchase reprints of this article please e-mail: [email protected]
Or visit our web site for further details: www.emeraldinsight.com/reprints
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
7
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Appendix
S
u
r
v
e
y
V
a
r
i
a
b
l
e
s
T
A
R
P
k
n
o
w
l
e
d
g
e
(
P
e
w
R
e
s
e
a
r
c
h
C
e
n
t
e
r
,
J
u
l
y
2
0
1
0
)
P
r
o
p
o
r
t
i
o
n
/
m
e
a
n
(
S
D
)
T
A
R
P
i
m
p
a
c
t
(
P
e
w
R
e
s
e
a
r
c
h
C
e
n
t
e
r
,
J
u
n
e
2
0
1
0
)
P
r
o
p
o
r
t
i
o
n
/
m
e
a
n
(
S
D
)
A
g
e
5
2
.
4
5
(
1
7
.
9
)
5
1
.
3
5
(
1
7
.
8
1
)
E
d
u
c
a
t
i
o
n
(
p
r
o
p
o
r
t
i
o
n
)
L
e
s
s
t
h
a
n
h
i
g
h
s
c
h
o
o
l
6
.
8
9
5
.
7
7
H
i
g
h
s
c
h
o
o
l
g
r
a
d
u
a
t
e
2
8
.
3
7
3
0
.
3
5
S
o
m
e
c
o
l
l
e
g
e
2
7
.
0
7
2
5
.
4
7
C
o
l
l
e
g
e
g
r
a
d
u
a
t
e
3
7
.
6
6
3
8
.
4
1
P
a
r
t
y
a
f
?
l
i
a
t
i
o
n
(
p
r
o
p
o
r
t
i
o
n
)
R
e
p
u
b
l
i
c
a
n
o
r
r
e
p
u
b
l
i
c
a
n
-
l
e
a
n
i
n
g
4
2
.
9
4
4
6
.
3
4
D
e
m
o
c
r
a
t
o
r
d
e
m
o
c
r
a
t
-
l
e
a
n
i
n
g
4
5
.
0
6
4
1
.
6
7
I
n
d
e
p
e
n
d
e
n
t
/
o
t
h
e
r
(
n
o
t
l
e
a
n
i
n
g
)
1
2
1
1
.
9
9
S
e
x
(
p
r
o
p
o
r
t
i
o
n
)
M
a
l
e
4
5
.
8
8
4
6
.
6
8
F
e
m
a
l
e
5
4
.
1
2
5
3
.
3
2
E
m
p
l
o
y
m
e
n
t
s
t
a
t
u
s
(
p
r
o
p
o
r
t
i
o
n
)
E
m
p
l
o
y
e
d
f
u
l
l
t
i
m
e
4
1
.
7
7
4
2
.
1
6
E
m
p
l
o
y
e
d
p
a
r
t
t
i
m
e
1
0
.
5
7
1
0
.
3
2
R
e
t
i
r
e
d
2
9
.
0
1
2
9
.
2
7
N
o
t
e
m
p
l
o
y
e
d
1
5
.
0
5
1
3
.
9
9
D
i
s
a
b
l
e
d
2
.
6
9
2
.
4
8
S
t
u
d
e
n
t
0
.
8
0
.
7
9
O
t
h
e
r
e
m
p
l
o
y
m
e
n
t
s
t
a
t
u
s
a
0
.
1
0
.
9
9
R
a
c
e
(
p
r
o
p
o
r
t
i
o
n
)
B
l
a
c
k
9
.
9
5
9
.
4
5
O
t
h
e
r
r
a
c
e
9
0
.
0
5
9
0
.
5
5
E
s
t
i
m
a
t
e
d
k
n
o
w
l
e
d
g
e
T
A
R
P
k
n
o
w
l
e
d
g
e
(
p
e
r
c
e
n
t
a
g
e
c
o
r
r
e
c
t
)
3
5
.
7
5
(
a
c
t
u
a
l
)
3
7
.
0
3
(
p
r
e
d
i
c
t
e
d
)
F
i
n
a
n
c
i
a
l
a
w
a
r
e
n
e
s
s
(
a
v
e
r
a
g
e
c
o
r
r
e
c
t
)
b
3
4
.
0
6
(
a
c
t
u
a
l
)
3
5
.
3
(
p
r
e
d
i
c
t
e
d
)
N
o
t
e
s
:
a
“
O
t
h
e
r
e
m
p
l
o
y
m
e
n
t
s
t
a
t
u
s
”
i
s
a
r
e
s
i
d
u
a
l
c
a
p
t
u
r
i
n
g
a
l
l
o
t
h
e
r
r
e
s
p
o
n
s
e
s
n
o
t
c
o
r
r
e
s
p
o
n
d
i
n
g
t
o
t
h
e
i
d
e
n
t
i
?
e
d
c
a
t
e
g
o
r
i
e
s
;
b
a
v
e
r
a
g
e
o
f
p
e
r
c
e
n
t
a
g
e
c
o
r
r
e
c
t
l
y
i
d
e
n
t
i
f
y
i
n
g
t
h
e
c
o
u
n
t
r
y
r
e
c
e
i
v
i
n
g
a
n
E
U
b
a
i
l
o
u
t
a
n
d
t
h
e
p
e
r
c
e
n
t
a
g
e
c
l
o
s
e
l
y
f
o
l
l
o
w
i
n
g
n
e
w
s
a
b
o
u
t
?
n
a
n
c
i
a
l
r
e
g
u
l
a
t
o
r
y
r
e
f
o
r
m
;
p
e
r
c
e
n
t
a
g
e
o
f
r
e
s
p
o
n
d
e
n
t
s
a
r
e
r
e
p
o
r
t
e
d
,
e
x
c
e
p
t
i
n
t
h
e
c
a
s
e
o
f
c
o
n
t
i
n
u
o
u
s
v
a
r
i
a
b
l
e
s
(
a
g
e
)
i
n
w
h
i
c
h
c
a
s
e
s
t
a
n
d
a
r
d
d
e
v
i
a
t
i
o
n
s
a
r
e
a
l
s
o
r
e
p
o
r
t
e
d
i
n
p
a
r
e
n
t
h
e
s
e
s
Table AI.
Summary statistics
JFEP
5,3
312
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
4
7
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
doc_917779233.pdf