Community Bank Establishment and Consumption Growth:

Evidence from Panel Study of Income Dynamics in USA

LI Yan
The Chinese University of Hong Kong

December 2024

Abstract

Consumption is a primary source of economic growth and key indicator of poverty. The es-
tablishment of community banks can provide credit resources, unlocking household consumption
potential and playing a crucial role in economic development. This study explores the role of
community banks in promoting consumption by using data from the Panel Study of Income Dy-
namics (PSID) for 11 waves from 1980 to 1990, and constructing a fixed-effects model using the
time-varying difference-in-differences (DID) method. The findings indicate that the establishment
of community banks effectively stimulates growth in local household consumption, primarily by
increasing household income and reducing precautionary savings. Therefore, both government and
financial institutions should further promote the development of regional financial institutions and
credit tools to promote economic growth.
Keywords: Community Bank, Credit Access, Consumption Growth

1 Introduction
Influenced by adverse conditions like war and deglobalization, the global economy has faced pressures
such as demand contraction, supply shocks, and weakened expectations in recent years. In this context,
unlocking consumption potential, improving financial services, and expanding demand have become
crucial for promoting economic development. Additionally, consumption is considered an essential
measure of poverty (Meyer and Sullivan, 2012). Promoting consumption growth is beneficial for
alleviating poverty and narrowing income disparities.
The relationship between the establishment of financial institutions and economic growth has been
a long-standing topic of interest and debate in economics. Community banks, which were once main-
stream financial institutions in the United States, play a significant role in the U.S. financial system
and have thus received considerable attention. Characterized by total assets of less than $1 billion
and a focus on serving local businesses and residents, community banks can enhance household income
through credit support, reduce the willingness to save, and consequently promote consumption.

1
 Community banks play a vital role in the U.S. financial system, and there is considerable research
on this topic. First, existing studies primarily focus on community banking operational issues, such
as competition (Marquis and Lounsbury, 2007), diversification (Stiroh, 2004), and mergers (Hughes
et al., 2019). Additionally, the impact of community banks on small and medium-sized enterprises
(SMEs) is also a key area of focus, with past research covering topics such as debt financing (Jagtiani
et al., 2016), bank-enterprise relationships (Berger et al., 2014), and loan growth (Avery and Samolyk,
2004). Finally, in developing countries like China and Nigeria, the role of community banks in poverty
alleviation and economic growth has been widely discussed (Lu et al., 2021; Ayadi et al., 2008).
However, research on the impact of community banks on American households and the economy is
quite limited, and their role warrants further exploration.
The research sample from the 1980s in the United States holds particular significance. On one hand,
as one of the wealthiest countries in the world, the U.S. has a highly developed financial sector. The
contributions of small financial institutions in the U.S. can serve as a practical reference for developing
countries and impoverished regions seeking to promote economic growth and escape poverty. On the
other hand, while community banks are no longer mainstream financial institutions in contemporary
America, the U.S. can draw upon past conclusions to reinforce its support for community banks,
potentially revitalizing the economy during current challenging times. Lastly, although many countries
have small financial institutions similar to community banks, data on these entities is relatively scarce,
particularly in developing countries. Using the U.S. as a sample allows for collecting large-scale data,
addressing the limitations of previous research that often relied on case studies and leading to more
universally applicable conclusions.
To address the above issues and conditions, based on the theories of credit access, liquidity con-
straints, and precautionary savings, this paper analyzes the relationship between the establishment of
community banks and household consumption growth using a sample of U.S. households participating
in the Panel Study of Income Dynamics (PSID) from 1980 to 1990. The findings indicate that the
establishment of community banks can alleviate household liquidity constraints, reduce precautionary
savings, and release consumption potential. Furthermore, we conducted a series of robustness checks
and validated the transmission mechanisms related to income and savings.
This paper contributes to bridging the research gap. First, while existing research has primarily
focused on the operational issues of community banks or their impact on small and medium-sized
enterprises, our study supports the positive effects of community bank establishment on households,
thereby enriching the relevant literature. Second, among similar studies, this paper provides a detailed
examination of the impact pathways of community banks both theoretically and empirically for the
first time.
The paper is organized as follows: the first section is an introduction, followed by a literature
review. The third section describes the sample and the research model. The fourth section presents
and discusses the results. The fifth section highlights the conclusions and practical significance and
recognizes the limitations.

2
2 Literature Review
From the late 1970s to the early 1980s, the strict restrictions on interstate banking competition in the
United States created a favorable environment for the operations of local small banks. Furthermore,
the Federal Reserve’s adjustments to monetary policy, which started in 1979, further increased public
demand for credit from small financial institutions. Community banks emerged in this macroeconomic
environment. Although there is no strict definition of community banks, following previous research,
this paper defines them as banks with assets of less than $1 billion, operating solely within their state,
and providing a full range of financial services.
Community banks play a crucial role in the functioning of the U.S. financial system and the broader
economy. Due to issues such as information asymmetry, small businesses and individuals often struggle
to obtain long-term financing (Udell et al., 1993). Community banks, constrained by statutory lending
limits, can focus on providing loans to small businesses and residents. In 1980, community banks
allocated 20% to 30% of their loan portfolios to commercial loans. In summary, the primary clients of
community banks are local SMEs and residents (De Young et al., 2004), providing essential banking
services in small towns and rural communities across the nation.
By concentrating on local financial markets, community banks can enhance access to credit in their
areas. Previous research has demonstrated that credit support is critical in creating job opportunities,
expanding production, and increasing welfare (Luan and Bauer, 2016). Credit is also essential for alle-
viating poverty and increasing income (Guirkinger, 2008). Therefore, the establishment of community
banks increases household income by providing credit, which in turn boosts consumption.
Then, standard macroeconomic models indicate that uncertainty plays a significant role in con-
sumption and savings decisions, as evidenced by the convexity of the marginal utility of consumption.
An increase in uncertainty generates positive additional savings, known as “precautionary savings”
(Leland, 1968; Sandmo, 1970). While precautionary savings can act as a buffer for households facing
unexpected economic pressures, they also crowd out household consumption and further restrict liquid-
ity. The credit provided by the establishment of community banks can enhance financial security for
households, allowing them to rely on credit to cope with sudden economic challenges, thereby reducing
the need for precautionary savings and increasing consumption.
Liquidity constraints are considered one of the main reasons why the permanent income model fails
to explain household consumption behavior (O. P. Attanasio and Weber, 2010). In imperfect credit
markets, households are unable to convert expected income into current consumption through credit
Zeldes, 1989). Families anticipating an increase in income are forced to delay subsequent consumption
growth Deaton, 1991). Therefore, the establishment of community banks can provide credit support,
alleviating liquidity constraints for local households, which enables smoother consumption patterns
and raises current consumption levels.
In conclusion, the theoretical framework supports the notion that increasing income and reducing
saving can lead to increased consumption. This paper argues that the establishment of community
banks can promote consumption growth among local households.

3
3 Methodology

3.1 Sample Selection and Data Source

To examine the relationship between the establishment of community banks and consumption growth,
this study selects families who participated in the Panel Study of Income Dynamics (PSID) survey from
1980 to 1990 as the sample. The relevant data on family consumption and characteristics all come from
the PSID. Staring from 1968, PSID is the longest-running longitudinal household survey in the world,
which collects data from over 5,000 families on topics including employment, income, expenditures,
health, and many other areas, providing an excellent national representation. The establishment year
of community banks in each state is sourced from the Federal Deposit Insurance Corporation (FDIC),
which is an independent agency created by Congress to maintain the stability of the nation’s financial
system. To ensure the validity of the research data, the following methods were employed for data
screening and processing: 1) Hawaii and U.S. overseas territories were excluded; 2) Samples with
household heads aged over the U.S. retirement age (66 years old) were excluded; 3) Samples with
missing values and outliers were excluded. 4) Only families where the head of the family remained
unchanged were retained to facilitate tracking. 5) Samples with zero consumption were excluded. The
final sample consists of 41,374 observations from 5,479 families.

3.2 Variable Setting

Consumption (Com): The logarithm of Consumption is the dependent variable in this study. Prior to
1999, the PSID had limited surveys related to consumption. This paper draws on the previous research
(Hall and Mishkin, 1982; O. Attanasio and Pistaferri, 2014), using food and rental expenditures
as a proxy for overall household consumption. In the PSID, food expenditures are composed of
expenditures at home, away from home, and food stamps. Given that food stamps are provided by
the government and are less influenced by consumption preferences, this study only calculates the sum
of food expenditures at home and away from home. Then, considering the PSID did not survey food
expenditures in 1988 and 1989, we used the interpolation method to fill in the data for these two
years. Additionally, we took 6% of the house’s value as rental expenditure and added that to the food
expenditure. Finally, We deflated the sample’s consumption data based on the U.S. 1980 CPI.
Community Bank (Policy): The establishment of community banks is the independent variable in
this study. While the definition of community banks is somewhat ambiguous, it generally refers to
banks with assets under $1 billion. Furthermore, to ensure a more accurate estimation of the impact
of community bank establishment on local consumption growth, we selected the establishment dates
of the first community bank in each state from the FDIC. We ensured that these community banks
operated solely within their respective states.
By drawing on previous research on consumption and the PSID (Gruber, 1998; Meyer and Mok,
2019), we selected several control variables to enhance the study’s validity. The control variables

4
mainly come from two levels: 1) Head level: the gender of the household head; 2) Household level:
household size, household income, and average education level of head and wife.
The explanation and set of variables are in Table 1.

Table 1: Variables Description

Symbol Variable Variable Definition

Com Total consumption Annual food and rental expenditure

Set to 1 if the state has already established

Policy Community Bank Establishment
the first community bank; otherwise 0.

Edu Education level Average education level of the head and wife

Sex Gender of head Which is set to 1 if head is male, otherwise 0

Income Total income Actual family total income

Size Family size Total members of the family

IND Individual-fixed effect Family dummies

STATE State-fixed effect State dummies

YEAR Year-fixed effect Year dummies

3.3 Methodological Remarks

To examine the relationship between the establishment of community banks and consumption growth,
this study employs a time-varying difference-in-differences model with individual, state and year-fixed
effects to construct Equation (1):

Log(Consumptionj,c,t) = β0 + β1Policyc,t + β2Controlsj,c,t + ωj + ϵc + ϕt + µj,c,t (1)

In this equation, Consumptionj,c,t is the food and rental consumption of the family j from state

c in year t, and we use its logarithm value. The variable Policyc,t is set to 1 if in year t the first
community bank has been established in state c. The term Controlsj,c,t represents the control variables
at two different levels, including educational level, gender, family income, and family size. ωj, ϵc and
ϕt are individual(family)-fixed effects, state-fixed effects and year-fixed effects. µj,c,t is residuals. β1
represents the effect of the establishment of a community bank on consumption. All data were clustered
at the family-level. Detailed definitions of these variables are given in Table 1.

5
4 Results and Discussions

4.1 Descriptive Statistics

Descriptive statistics are shown in Table 2, which summarizes the distribution and characteristics of
each variable analyzed in the study. The results show that the Mean of consumption and Log(consumption)
are 4809 and 8.124, and the Mean of Policy is 0.438. The mean close to 0.5 indicates no significant dif-
ference in the scale of observations before and after the establishment of the bank. The maximum and
minimum values of consumption are 0.827 and 51,490, respectively, indicating a significant difference
in sample distribution and a high degree of differentiation. The same applies to the data on income.

Table 2: Descriptive Statistics

Variable Name Obs Mean SD Min Max

Com 41,374 4,809 3,895 0.827 51,490

Log(Com) 41,374 8.124 0.917 -1.189 10.849
Policy 41,374 0.438 0.495 0 1
Income 41,374 27,002 31,149 0 1,412,200
Edu 41,374 3.622 1.890 0 8
Sex 41,374 0.760 0.427 0 1
Size 41,374 2.946 1.541 1 14

4.2 Regression Result Analysis

Based on Equation 1, we systematically and progressively incorporated control variables from two
distinct levels of analysis, with the comprehensive results presented in columns (1) to (3) of Table 3.
This methodological approach allows us to better understand and illustrate the nuanced impact of the
establishment of community banks on local consumption patterns.
In columns (1) through (3), the coefficients for the variable ”Policy” consistently demonstrate
positive values and are statistically significant at the 1% level. This strong statistical significance
indicates that the establishment of community banks has a substantial and positive effect on promoting
local consumption growth. The empirical findings we observed not only reinforce our theoretical
inferences but also highlight the critical role that community banks play in enhancing economic activity
within the region.
Focusing specifically on the results presented in column (3), we find that the annual food expendi-
ture of households in the state is projected to increase by an average of 3.9% following the establishment
of community banks. This increase in food expenditure reflects a broader trend of enhanced consump-
tion growth.

6
 Table 3: Regression Results

VARIABLES (1) (2) (3)

Policy 0.047*** 0.043*** 0.039***

(3.57) (3.36) (3.26)
Sex 0.348*** 0.311***
(10.06) (10.10)
Edu 0.128*** 0.051***
(20.21) (7.93)
Income 0.000***
(5.04)
Size 0.154***
(24.56)

Indivi.-Fixed effect Yes Yes Yes

State-Fixed effect Yes Yes Yes
Year-Fixed effect Yes Yes Yes
Constant 8.122*** 7.397*** 7.168***
(1,433.10) (283.65) (279.41)
Observations 41,374 41,374 41,374
R-squared 0.736 0.757 0.776
Notes: t-statistics in parentheses *** p is less than 0.01, ** p is less than 0.05, * p is less than 0.1

4.3 Robustness Test: Parallel Test

To investigate the dynamic changes in consumption before and after the establishment of community
banks and to ensure that the promoting effect of community bank establishment on consumption
growth does not stem from pre-existing trends, this study draws on the methodologies of previous
research (Jacobson et al., 1993; Kudamatsu, 2012). We employ an event study approach to conduct
a parallel trends test, thereby enhancing the robustness of the results. Specifically, we set up the
regression equation as follows:
Σ
3
k
δkD c,t
Log(Consumptionj,c,t ) = α + + γControlsj,c,t + ωj + ϵc + ϕt + µj,c,t (2)
k=−5
k̸=0

In Equation (2), let the year that state c establishes the first community bank be s, and D is a dummy
variable indicating the establishment of the community bank. When t − s = k, Dc,t
k = 1; otherwise,

k
D c,t = 0. To maintain a reasonable period, we define D3c,t = 1 when t − s > 3 and Dc,t
−5 = 1 when

t − s < −5. The control variables, fixed effects, and residuals are the same as in Equation (1).
To avoid multicollinearity, this study uses the year of the establishment as the baseline year,

7
excluding the dummy variable for k = 0. Additionally, following the previous approach (Beck et al.,
2010), this study performed de-trending and centering in the parallel trends test. The final results of
the parallel trends test are illustrated in the Table 4 and Figure 1 below:

Table 4: Parallel Test Regression Results

VARIABLES (1)

Pre 5 0.009
(0.60)
Pre 4 -0.003
(-0.84)
Pre 3 -0.006
(-1.25)
Pre 2 -0.016
(-2.19)
Pre 1 0.026
(0.39)
Post 1 0.058**
(2.43)
Post 2 0.078***
(2.89)
Post 3 0.099***
(2.97)
Baseline Year Omitted

Control Variables Yes

Indivi.-Fixed effect Yes
State-Fixed effect Yes
Year-Fixed effect Yes
Constant 7.172***
(264.20)
Observations 41,374
R-squared 0.776
Notes: t-statistics in parentheses *** p is less than 0.01, ** p is less than 0.05, * p is less than 0.1

As Table 4 and Figure 1 shows, in the pre-baseline period, the 95% confidence intervals all include 0,
indicating that the explanatory variables’ regression coefficients and trends are insignificant. However,
in the post-baseline period, the regression coefficients of the explanatory variables are significantly
positive and maintain an upward trend. This indicates that after the establishment of community
banks, local household consumption significantly increases, confirming that the consumption growth

8
is indeed driven by the establishment of the banks rather than a time trend effect.

Figure 1: Parallel Test

4.4 Robustness Test: Placebo Test

To rule out the influence of other concurrent events or unobserved characteristics of states, we con-
ducted a placebo test. Specifically, we randomly designed the year of establishment of community
banks in the sample and the corresponding region, thereby generating a dummy variable for the policy
as a new explanatory variable in the regression analysis, and then performed the regression. This
process was repeated 500 times to obtain the regression coefficients and density distribution of the
placebo test. If other concurrent events drove the growth of household consumption, we would expect
to see similar results after randomly assigning the establishment and location of community banks.
However, the findings presented in Figure 2 reveal that both the P-value and the density distri-
bution conform to a normal distribution symmetric at x=0. Notably, the actual coefficient from the
baseline regression, which is 0.039, exceeds all coefficients obtained from the placebo tests. In conclu-
sion, the outcomes of the placebo test strongly suggest that consumption growth is not influenced by
contemporaneous events or regional characteristics.

9
 Figure 2: Placebo Test

4.5 Robustness Test: Adjustment of sample and variables

In order to make the estimation more accurate and reliable, this paper specifically adjusts the samples
and variables and re-runs the regression.
First, considering that low-income households have a relatively strong demand for credit and a
higher marginal propensity to consume food, the establishment of community banks is likely to have
a more pronounced effect on the consumption of impoverished populations. Therefore, we restricted
the sample to households with annual incomes above the U.S. poverty line for the regression analysis.
The results are shown in column (1).
Second, due to the lack of food consumption data for 1988 and 1989 in the PSID, this study’s base-
line regression supplemented these two years using the interpolation method. To avoid any potential
impact of the supplemented data on the accuracy of the estimates, we conducted the regression using
only the original data (excluding 1988-1989). We replaced the consumption variable with only food
expenditures, ensuring that the dependent variable underwent sufficient adjustment. The results are
shown in column (2).
Third, due to the increase in interest rates in the late 1970s, the U.S. economy experienced a
prolonged negative growth period from early 1980 to the end of 1981. Therefore, we excluded the data

10
from 1980-1981 to ensure a relatively stable macroeconomic environment, and the results are shown
in column (3).
Finally, the baseline regression includes rental expenditure and food expenditure into consump-
tion. Considering that rental expenditure is highly sensitive to credit access, we only include food
expenditure into consumption calculations. The results are presented in column (4).
The four robustness checks’ results, conducted through variables and sample adjustments, are
presented in Table 5. These findings indicate that the coefficient for Policy consistently remains
positive and is significant at the 1% or 5% level, thereby affirming that the promotion of consumption
attributable to the establishment of community banks is robust across various specifications.

Table 5: Robustness Test Regression

Variables (1) (2) (3) (4)

Policy 0.026** 0.027*** 0.034*** 0.029**

(2.39) (2.15) (2.87) (2.48)
Control Variables Yes Yes Yes Yes

Indivi.-Fixed effect Yes Yes Yes Yes

State-Fixed effect Yes Yes Yes Yes
Year-Fixed effect Yes Yes Yes Yes
Constant 7.313*** 7.167*** 7.225*** 7.262***
(246.42) (324.36) (250.81) (338.82)
Observations 32,234 35,028 33,120 41,279
R-squared 0.787 0.678 0.798 0.690
Notes: t-statistics in parentheses *** p is less than 0.01, ** p is less than 0.05, * p is less than 0.1

5 Analysis of Mechanisms
This study argues that the establishment of community banks promotes household consumption
through two pathways: increasing household income and reducing household precautionary savings.
Given the clear impact and strong causal relationship of income and savings on consumption, we draw
on previous econometrics research (Jiang, 2022) to examine the effects of community bank establish-
ment on household income and savings separately. This approach aims to verify whether these two
factors serve as channels of influence.
First, we continue to utilize the PSID survey data, with household income as the dependent variable;
the results are shown in column (1). Secondly, since the PSID only surveyed savings decisions in 1984
and 1989 from 1980 to 1990, we used data from these two years, with the results in column (2).
In column (2), the savings decision is a binary variable: it equals one if the household responds

11
affirmatively to both general savings (cash, bank accounts) and other savings (checking accounts,
insurance, pensions, etc.); otherwise, it equals 0. The results in Table 6 indicate that the establishment
of community banks increases total household income while also reducing the likelihood of households
making savings decisions, and the results are significant at 5% level.

Table 6: Regression of Impact Mechanisms

Variables (1) Income (2) Saving Decision

Policy 944.169** -0.236**

(2.43) (-2.31)
Control Variables Yes Yes

Indivi.-Fixed effect Yes Yes

State-Fixed effect Yes Yes
Year-Fixed effect Yes Yes
Constant 81.67 0.171***
(0.60) (2.77)
Observations 43,354 5,964
R-squared 0.677 0.640
Notes: t-statistics in parentheses *** p is less than 0.01, ** p is less than 0.05, * p is less than 0.1

6 Discussion and Conclusion

The scale and number of community banks in the United States have significantly contracted un-
der the influence of various factors such as policy restrictions, intensified competition, and financial
digitalization. However, research on community banks remains highly relevant. This paper utilizes
PSID surveys to obtain the establishment dates of community banks across various states. It draws
on representative data regarding household consumption and income in the U.S. from 1980 to 1990
via the PSID to study the impact of community bank establishment on consumption growth. The
findings indicate that the establishment of community banks can significantly promote consumption
growth among local households. Subsequently, the paper strengthens the robustness of its conclusions
through parallel tests, placebo tests, and sample modifications.
Finally, this study explores the mechanisms through which community banks affect consumption.
It finds that, on the one hand, the establishment of community banks provides credit support, promot-
ing economic growth by creating jobs and expanding production, thereby increasing residents’ income
and consumption. On the other hand, household uncertainty may encourage families to increase their
expected savings; however, the access to credit provided by community banks can enhance house-
holds’ ability to cope with sudden economic pressures, reducing the need for savings and releasing
consumption potential. Furthermore, liquidity constraints make it difficult for households to convert

12
expected income into current consumption, which contradicts the permanent income model. At the
same time, the establishment of community banks can improve household liquidity, thereby smoothing
and increasing consumption.
This paper has policy formulation, financial institution governance, and poverty alleviation im-
plications. First, given the significant reduction in the number of community banks in the U.S., it
may be worthwhile to consider policies that promote the development of community banks to alleviate
economic pressures in the face of challenges. Second, impoverished regions or developing countries
can draw lessons from the U.S. experience by using community banks to provide credit support for
poverty alleviation efforts. Lastly, community banks can further enhance their support for local SMEs
and low-income families, thereby maximizing their social value. Community banks might also consider
expanding their customer outreach through digitalization and other reform measures.
The implicit problem of this paper is that the period of community banks establishment coincides
with the U.S. savings and loan crisis. In the early 1980s, the special macroeconomic environment,
policies and regulations gave rise to the rapid development of small financial institutions. The growth
of community banks and S&Ls during this period was largely due to their issuance of bad loans and
disorderly expansion without reasonable supervision. By the end of the 1980s, policy shifts and early
risk accumulation led to a large number of bankruptcies of such institutions (Curry and Shibut, 2000;
Kane, 1989). In our study, we studied the impact of the first community bank established in each
state on consumption, but did not take into account that its impact on consumption may come from
the special background of the times. In other words, its impact on consumption may not be due to the
corresponding theoretical and mechanism analysis, and it may not be persistent with the bankruptcy
of financial institutions. This problem affects the accuracy of the conclusion and weakens the reference
significance of the research to other regions and the present.
Our study still has some other limitations. First, the lack of a unified definition for community
banks complicates the identification of the first community bank in various regions, and inaccuracies
in the establishment dates of community banks may affect the conclusions. Second, our research
primarily focuses on data from the U.S. between 1980 and 1990, and given the substantial decrease in
the number of community banks today, the practical significance of this paper is somewhat limited.
Finally, while the paper finds that community banks hold additional significance for impoverished and
developing countries, it does not conduct an in-depth study on this aspect.
Future research could explore the following areas: First, topics could be closer to current events,
such as studying the negative impacts of community bank closures and mergers on consumption.
Second, a deeper investigation could be into their effects on poverty alleviation and other household
economic indicators. Third, the research scope could be further expanded to developing countries
to seek more general conclusions. Fourth, More detailed consumption data than just food and rent
expenditures can be used for further research, such as Consumer Expenditure Surveys. Lastly, the
study should try to change the sample year to eliminate the impact of the U.S. savings and loan crisis
at the same time on the robustness of the results.

13
References
Attanasio, O., & Pistaferri, L. (2014). Consumption inequality over the last half century: Some evidence
using the new psid consumption measure. American Economic Review, 104 (5), 122–126. https:
//doi.org/10.1257/aer.104.5.122
Attanasio, O. P., & Weber, G. (2010). Consumption and saving: Models of intertemporal allocation and
their implications for public policy. Journal of Economic Literature, 48 (3), 693–751. https:
//doi.org/10.1257/jel.48.3.693
Avery, R. B., & Samolyk, K. A. (2004). Bank consolidation and small business lending: The role of
community banks. Journal of Financial Services Research, 25, 291–325. https://doi.org/10.
1023/B:FINA.0000020667.29802.1c
Ayadi, O. F., Hyman, L. M., & Williams, J. (2008). The role of community banks in economic devel-
opment: A nigerian case study. Savings and Development, 159–173. https://www.jstor.org/
stable/25830975
Beck, T., Levine, R., & Levkov, A. (2010). Big bad banks? the winners and losers from bank dereg-
ulation in the united states. The Journal of Finance, 65 (5), 1637–1667. https://doi.org/10.
1111/j.1540-6261.2010.01589.x
Berger, A. N., Goulding, W., & Rice, T. (2014). Do small businesses still prefer community banks?
Journal of Banking & Finance, 44, 264–278. https://doi.org/10.1016/j.jbankfin.2014.03.016
Curry, T., & Shibut, L. (2000). The cost of the savings and loan crisis: Truth and consequences. FDIC
Banking Review, 13 (2). https://elischolar.library.yale.edu/ypfs-documents/11175/
De Young, R., Hunter, W. C., & Udell, G. F. (2004). The past, present, and probable future for
community banks. Journal of Financial Services Research, 25, 85–133. https://doi.org/10.
1023/B:FINA.0000020656.65653.79
Deaton, A. (1991). Saving and liquidity constraints. Econometrica, 59 (5), 1221. https://doi.org/10.
2307/2938366
Gruber, J. (1998). Unemployment insurance, consumption smoothing, and private insurance: Evidence
from the psid and cex. Research in Employment Policy, 1 (1), 3.
Guirkinger, C. (2008). Understanding the coexistence of formal and informal credit markets in piura,
peru. World Development, 36 (8), 1436–1452. https://doi.org/10.1016/j.worlddev.2007.07.002
Hall, R. E., & Mishkin, F. S. (1982). The sensitivity of consumption to transitory income: Estimates
from panel data on households. Econometrica, 50 (2), 461–481. https://doi.org/10.2307/
1912638
Hughes, J. P., Jagtiani, J., Mester, L. J., & Moon, C. (2019). Does scale matter in community bank
performance? evidence obtained by applying several new measures of performance. Journal of
Banking & Finance, 106, 471–499. https://doi.org/10.1016/j.jbankfin.2019.07.005
Jacobson, L. S., LaLonde, R. J., & Sullivan, D. G. (1993). Earnings losses of displaced workers. The
American Economic Review, 83 (4), 685–709. http://www.jstor.org/stable/2117574

14
Jagtiani, J., Kotliar, I., & Maingi, R. Q. (2016). Community bank mergers and their impact on small
business lending. Journal of Financial Stability, 27, 106–121. https://doi.org/10.1016/j.jfs.
2016.10.005
Jiang, T. (2022). Mediating effects and moderating effects in causal inference. China Industrial Eco-
nomics, 5, 100–120. https://doi.org/10.19581/j.cnki.ciejournal.2022.05.005
Kane, E. J. (1989). The s&l insurance mess: How did it happen? The Urban Institute Press. https:
//archive.org/details/slinsurancemessh0000kane/mode/2up
Kudamatsu, M. (2012). Has democratization reduced infant mortality in sub-saharan africa? evidence
from micro data. Journal of the European Economic Association, 10 (6), 1294–1317. https:
//doi.org/10.1111/j.1542-4774.2012.01092.x
Leland, H. E. (1968). Saving and uncertainty: The precautionary demand for saving. The Quarterly
Journal of Economics, 82 (3), 465–473. https://doi.org/10.2307/1879518
Lu, Z., Wu, J., Li, H., & Nguyen, D. K. (2021). Local bank, digital financial inclusion and sme financing
constraints: Empirical evidence from china. Emerging Markets Finance and Trade, 58 (6),
1712–1725. https://doi.org/10.1080/1540496X.2021.1923477
Luan, D. X., & Bauer, S. (2016). Does credit access affect household income homogeneously across
different groups of credit recipients? evidence from rural vietnam. Journal of Rural Studies,
47, 186–203. https://doi.org/10.1016/j.jrurstud.2016.08.001
Marquis, C., & Lounsbury, M. (2007). Vive la r´esistance: Competing logics and the consolidation of
us community banking. Academy of Management Journal, 50 (4), 799–820. https://doi.org/
10.5465/amj.2007.26279172
Meyer, B. D., & Mok, W. K. (2019). Disability, earnings, income, and consumption. Journal of Public
Economics, 171, 51–69. https://doi.org/10.1016/j.jpubeco.2018.06.011
Meyer, B. D., & Sullivan, J. X. (2012). Identifying the disadvantaged: Official poverty, consumption
poverty, and the new supplemental poverty measure. Journal of Economic Perspectives, 26 (3),
111–136. https://doi.org/10.1257/jep.26.3.111
Sandmo, A. (1970). The effect of uncertainty on saving decisions. The Review of Economic Studies,
37 (3), 353–360. https://doi.org/10.2307/2296725
Stiroh, K. J. (2004). Do community banks benefit from diversification? Journal of Financial Services
Research, 25, 135–160. https://doi.org/10.1023/B:FINA.0000020657.59334.76
Udell, G. F., Carey, M., Prowse, S. D., & Rea, J. D. (1993). The economics of the private placement
market. Staff Studies, 80, 5. https://www.researchgate.net/publication/5039967
Zeldes, S. P. (1989). Consumption and liquidity constraints: An empirical investigation. Journal of
Political Economy. https://doi.org/10.1086/261605

15