REVISTA DE ECONOMÍA MUNDIAL 68, 2024, 141-166

ISSN: 1576-0162

DOI: http://dx.doi.org/10.33776/rem.v0i68.7996

DETERMINANTS OF THE PROFITABILITY OF SAVINGS BANKS IN THE

US AND THE MODERATING EFFECT OF THE PANDEMIC CRISIS

DETERMINANTES DE LA RENTABILIDAD DE LAS CAJAS DE AHORRO EN

ESTADOS UNIDOS Y EL EFECTO MODERADOR DE LA CRISIS PANDÉMICA

Marco Amaral

mamaral@ipca.pt

Polytechnic Institute of Cávado and Ave

Recibido: octubre 2023; aceptado: julio 2024

ABSTRACT

The pandemic crisis that began in 2020 and lasted until 2021 has raised

concerns about the impact on banking proﬁtability. As a result, the global

banking system suffered a drop in the proﬁtability of its banking business as

part of the crisis. Taking into account the problem of proﬁtability in the banking

sector, some studies have been increasingly highlighted both in academia and

in the ﬁnancial market. Therefore, this study aims to analyze the determinants

of bank proﬁtability for banks operating in the savings segment in the United

States, for the periods between 2012 and 2022. To this end, four factors

explaining proﬁtability (default rate, liquidity, solvency and productivity) were

analyzed for 57 banks. In addition, we wanted to assess the moderating effect

that the pandemic crisis had on the banking sector. To obtain the results, a

panel data analysis model was used, combining cross-section data (banks)

and time-series data (years), and a strongly balanced panel was obtained.

Four multiple linear regression models were then used, and it was possible to

identify liquidity and solvency as the main positive factors in bank proﬁtability.

In contrast, the factors of default rate and productivity have a negative inﬂuence

on bank proﬁtability. The results also show that banks with high default rates

had greater difﬁculties during the period of the pandemic crisis, that is, the

crisis negatively moderated the effect of default on proﬁtability.

Keywords: Proﬁtability, pandemic crisis, banks, US, panel data.

RESUMEN

La crisis pandémica que comenzó en 2020 y duró hasta 2021 ha suscitado

preocupación por su impacto en la rentabilidad bancaria. Como consecuencia,

el sistema bancario mundial sufrió una caída de la rentabilidad de su negocio

bancario en el marco de la crisis. En vista del problema de la rentabilidad en

el sector bancario, se ha hecho cada vez más hincapié en algunos estudios

tanto en el mundo académico como en el mercado ﬁnanciero. Por lo tanto,

este estudio tiene como objetivo analizar los determinantes de la rentabilidad

bancaria para los bancos que operan en el segmento de ahorro en los Estados

Unidos de América, para los períodos comprendidos entre 2012 y 2022.

Para ello, se han analizado cuatro factores explicativos de la rentabilidad

(tasa de morosidad, liquidez, solvencia y productividad) para 57 bancos.

Además, se trató de evaluar el efecto moderador que tuvo la crisis pandémica

en el sector bancario. Para obtener los resultados, se utilizó un modelo

de análisis de datos de panel que combina datos de sección transversal

(bancos) y datos de serie temporal (años), obteniendo un panel fuertemente

equilibrado. A continuación, se utilizaron cuatro modelos de regresión lineal

múltiple y fue posible identiﬁcar la liquidez y la solvencia como los principales

factores positivos de la rentabilidad bancaria. Por el contrario, los factores

de morosidad y productividad inﬂuyen negativamente en la rentabilidad

bancaria. Los resultados también muestran que los bancos con altas tasas

de morosidad tuvieron mayores diﬁcultades durante el periodo de la crisis

pandémica, es decir, la crisis moderó negativamente el efecto de la morosidad

en la rentabilidad.

Palabras clave: Rentabilidad; crisis pandémica; bancos; datos de panel.

JEL Classiﬁcation/ Clasiﬁcación JEL: G01, G21, G28.

REVISTA DE ECONOMÍA MUNDIAL 68, 2024, 141-166

1. INTRODUCTION

Given the recent adverse economic contexts (ﬁnancial crisis and pandemic

crisis) in which banks operate, it is becoming increasingly urgent to understand

the factors that may determine their proﬁtability and the explanatory variables

that may inﬂuence banking performance. Recently, a large number of studies

on this subject (Messai et al., 2015; Duraj and Moci, 2015, Titko et al.,

2015; Elisa and Guido, 2016; Alshatti, 2016; Le and Ngo, 2020), among

others, have shown a set of dominant factors that drive proﬁtability in the

banking sector. For the authors Wahdan and Leithy (2017), the factors that

affect bank proﬁtability can come from external or internal sources. In the

same line of research, Serwadda (2018) states that external factors include

the government’s macroeconomic policies, central bank interest rates, climate

change and the current COVID-19 pandemic. In turn, internal factors include

interest income, overdue credit, capital adequacy, asset size and others (Islam

et al., 2017). Thus, the study of the determinants of bank proﬁtability proves

to be fruitful and timely.

The purpose of this paper is to analyze the determinants of the proﬁtability

of savings banks in the US and to assess the moderating effect that the

pandemic crisis has had on this sector. To achieve this goal, 57 savings banks in

the US banking system were analyzed for the period 2012 to 2022. In order to

analyze the proﬁtability of US savings banks, two bank proﬁtability indicators

were used and related to a set of management indicators, separated into four

categories, namely: level of default, liquidity, solvency and productivity. These

indicators make it possible to work with a signiﬁcant number of variables such

as costs, revenues, loans, deposits, assets, capital, employees, among others,

and relate them to their respective proﬁtability indicators, measured by RoA

(Return on Assets) and RoE (Return on Equity). In addition, in order to measure

the functional relationships between the variables in the study, control variables

such as interest rates, the consumer price index, the political organization

system, and size were considered. Speciﬁcally, we strive to answer a set of

questions that allow us to validate the level of proﬁtability of savings banks

in the US banking system and identify the divergences between proﬁtability

and the internal and external characteristics of banks, such as: default rate,

liquidity, solvency, productivity, inﬂation, central bank reference rate, type of

government, and size. In addition, for the sample period considered, we looked

at whether the pandemic played a moderating role in these effects. In this

144 Marco Amaral

sense, in addition to focusing on the relationship between internal and external

factors and the proﬁtability of the banking sector, the debate in this research

also aims to assess the moderating effect of the pandemic crisis. Thus, it is

hoped that this study will make a strong contribution to research in this thematic

area, essentially in the analysis of the proﬁtability of US banking institutions,

thus making it possible to portray the reality of US banking specializing in the

savings segment, as well as contributing to the teachings of bank proﬁtability

focused on its internal and external determinants.

The ﬁnal document of this work is structured as follows: in addition to

the ﬁrst introductory section, a second section on the literature review is

presented, highlighting the evidence of similar studies and the formulation

of the hypotheses to be tested. This section describes the sample, data

collection, study variables, speciﬁcation of the econometric model and

statistical methods, the results obtained and their discussion. Finally, the

fourth point, the conclusion of the study, is presented, which reﬂects on the

main conclusions of this research, as well as the main limitations of the study.

2. RELEVANT LITERATURE AND DEVELOPMENT OF HYPOTHESES

This section presents the literature review of this study in order to ensure and

substantiate the purposes of the work carried out, as well as the development

of the hypotheses to be tested, through the hypotheses formulated for this

purpose.

2.1. LITERATURE REVIEW

There are a number of studies that relate measures of bank ﬁnancial

proﬁtability, particularly with regard to the factors that determine the

proﬁtability of credit institutions.

Analyzing the literature for the banking market on bank proﬁtability, the

works carried out by authors such as Athanasoglou et al. (2008), Sanchez et

al. (2017) and Bikker and Vervliet (2018), use the Return on Assets (RoA) and

Return on Equity (RoE) indices as the main indicators to measure results and

assess their evaluations. Therefore, to ensure the purposes of this study, a set

of data in different environments from various authors was analyzed. In this

way, it was possible to distinguish some studies insofar as the results obtained

are different, but which, despite presenting mixed associated relationships,

have in common combinations of internal and external determinants to explain

bank proﬁtability.

In Europe, Athanasoglou et al. (2008) analyzed the proﬁtability of Greek

commercial banks, measured by RoA and RoE, from 1985 to 2001, in the

light of the recent economic recession. These authors carried out the most

popular decomposition of the determinants of bank proﬁtability, adopting

three categories of factors that explain bank proﬁtability. First, through bank-

speciﬁc factors, using variables such as size, capital, credit risk, loans, revenue

145

DETERMINANTS OF THE PROFITABILITY OF SAVING S BANKS IN THE US AND THE MODERATING EFFECT OF THE PANDEMIC CRISIS

REVISTA DE ECONOMÍA MUNDIAL 68, 2024, 141-166

diversiﬁcation, type of bank and efﬁciency, they found a signiﬁcant positive

effect for the capital variable and a signiﬁcant negative effect for the credit

risk variable, while the results for bank size were insigniﬁcant. In a second

category of determinants of bank proﬁtability, they adopted industry/sector-

speciﬁc factors such as ownership and bank concentration and concluded

that there is no clear relationship between sector concentration and bank

proﬁtability. Finally, in the third category of determinants, they looked at the

macroeconomic environment (external factors) with a special focus on the

variable of economic growth measured by Gross Domestic Product (GDP), with

the authors concluding that strong economic growth combined with higher

interest rates is likely to increase bank proﬁtability.

In a cross-sectional analysis of 154 ﬁnancial entities from 22 countries for

the period 2005-2010, the study by Sanchez et al. (2017) aims to understand

the impact of corporate social performance on the ﬁnancial performance

of the banking sector. In this sense, the authors’ study consists of verifying

which dimensions of corporate social performance (corporate governance,

employee relations, community relations and product responsibility) had the

greatest impact on banks’ ﬁnancial proﬁtability (measured by RoA and RoE),

as well as whether the 2008 ﬁnancial crisis played a moderating role in these

effects. The authors also used control variables such as bank size and leverage

to avoid biased results. In order to assess the impact of social performance

on ﬁnancial proﬁtability, the authors used dynamic panel data through the

multiple linear regression equation for the ﬁxed effects model. The results

showed that banks with better employee relations and corporate governance

had better proﬁtability. However, the ﬁnancial crisis negatively moderated the

effect of corporate governance, suggesting ﬂaws in its mechanisms. Product

liability, contrary to the authors’ expectations, did not positively inﬂuence bank

proﬁtability. Finally, the authors analyzed the moderating effect of the ﬁnancial

crisis and concluded that the crisis has an interaction effect on the relationship

between social performance and banks’ ﬁnancial proﬁtability. This effect ﬂows

mainly through the variables of corporate governance and community.

On the other side of the continent, namely in the US banking system,

the study by the authors Bikker and Vervliet (2018), for the years 2001 to

2015, made it possible to consider an analysis of the period before and after

the ﬁnancial crisis. The work carried out by the authors aims to explore the

relationship between bank proﬁtability and the economic environment of low

interest rates. The sample includes a set of panel data for 3,582 commercial

and savings banks, using proﬁtability measures such as RoA, RoE, net interest

margin and proﬁt. As determinants, the authors selected a set of bank-speciﬁc

variables (size, loans, capital, credit risk and revenue diversiﬁcation) and

macroeconomic variables (GDP, inﬂation and interest rates). From the results

obtained, the authors concluded that variables such as capital, loans and size

have a signiﬁcant positive effect on bank proﬁtability, in contrast, the credit

risk variable has a negative effect and the macroeconomic variables GDP and

inﬂation were considered insigniﬁcant in explaining bank proﬁtability. Finally,

146 Marco Amaral

the authors conﬁrm that, in part, the low interest rate environment harms bank

proﬁtability and crushes net interest income.

Table 1 shows, by author, a number of other studies on the subject of bank

proﬁtability.

TABLE 1. STUDIES ON BANK PROFITABILITY, BY AUTHOR

Authors

(Year)

Period

(Sample) Proﬁtability Metrics Determinants Conclusion

Bourke

(1989)

1972-1981

(Banks from

12 countries)

RoA

Liquidity

Capital

Leverage

Ownership

Concentration

- The results show that banking

concentration has a positive

effect on proﬁtability. On the

other hand, capital and liquidity

show a negative relationship.

Demirguc-Kunt and

Huizinga

(1999)

1988-1995

(Banks in 80

countries)

Net Interest Margin

Dimension

Revenue

Capital

Liquidity

Credit Risk

GDP

Inﬂation

Interest Rates

- The authors note that deter-

minants such as capital, GDP

and inﬂation show a positive

and statistically signiﬁcant

association. In contrast, the

liquidity variable shows a

negative relationship with bank

proﬁtability.

Tregenna (2009) 1994-2005

(EUA)

RoA

RoE

Concentration

Market Share

Size

Efﬁciency

- The study shows that banking

concentration increases proﬁt-

ability, even when the largest

banks are excluded.

Fronk

(2016)

1985-2015

(EUA) RoA

GDP

Unemployment Rate

Interest RateSpread

- Macroeconomic factors are

responsible for proﬁtability be-

ing so low during and after the

ﬁnancial crisis.

Paroush and

Schreiber (2019)

1995-2015

(EUA) RoA Capital

Risk

The authors found that capital

is positively related to proﬁt-

ability, while risk is negatively

related.

Mendoza et al.

(2020)

1994-2011

(Banks from

134 coun-

tries)

RoA

Net Interest Margin

Concentration

Diversiﬁcation

Risk

Liquidity

Capital

Loans

Efﬁciency

GDP

Inﬂation

- Determinants such as

banking concentration, GDP,

diversiﬁcation of services and

products show a positive and

signiﬁcant relationship with

bank proﬁtability. On the other

hand, inﬂation shows a negative

relationship with proﬁtability.

Veeramoothoo and

Hammoudeh

(2022)

2010-2017

(EUA)

RoA

RoE

Financing

Liquidity

Size

Capital

Risk

GDP

- The authors found that small

banks are more vulnerable to

short-term liquidity risks and

large banks are more suscep-

tible to medium and long-term

liquidity risks.

Wang

(2023)

2008-2016

(Japan and

EUA)

RoA

Risk

Liquidity

Efﬁciency

Capital

Size

GDP

- The author’s study shows that

holdings of risky securities in

Japan have a positive effect,

while in the US it has a negative

effect on the proﬁtability of

large banks.

Source: Author´s own creation.

147

DETERMINANTS OF THE PROFITABILITY OF SAVING S BANKS IN THE US AND THE MODERATING EFFECT OF THE PANDEMIC CRISIS

REVISTA DE ECONOMÍA MUNDIAL 68, 2024, 141-166

The following subsections present the determinants that can inﬂuence

bank proﬁtability, and which therefore relate to bank proﬁtability. Thus, in light

of the literature, the following ﬁve hypotheses were developed, separated by

four speciﬁc categories for the sector of activity (default, liquidity, solvency

and productivity) and by a category designed to measure the effect of the

pandemic crisis on bank proﬁtability, as shown in Table 2.

TABLE 2. SUMMARY OF THE STUDY HYPOTHESES

Factor Argument Hypotheses

Default Default rate on the portfolio of loans

granted to customers

H1: The sluggishness of banks, as measured by the

default rate on the loan portfolio, negatively affects

banks proﬁtability

Liquidity Transforming customer funds into

customer loans

H2: Bank liquidity has a positive (negative) effect on

banks proﬁtability

Solvency Capital adequacy H3: Bank solvency has a positive (negative) effect on

bank proﬁtability

Productivity Employee production H4: Bank productivity has a positive effect on bank

proﬁtability

Pandemic Moderating effect of the pandemic

crisis - COVID19

H5: The moderating effect of the pandemic crisis has a

signiﬁcant inﬂuence on the relationship between the four

factors and bank proﬁtability

Source: Author´s own creation.

2.2. DEFAULT

The default variable is intended to reﬂect the sluggishness of a bank’s

customer loan portfolio. In order to analyze the default rate of the loan portfolio,

this study used the effect of impairments and loan provisions on the customer

loan portfolio, thus making it possible to verify the weight of impairments

recorded in each ﬁnancial year in the loan portfolio, i.e. in total loans. There is

a general consensus among the authors, Demirguc-Kunt and Huizinga (1999),

DeYoung and Rice (2004), Athanasoglou et al. (2008), Kosmidou (2008), Liang

et al. (2013), Barata (2014), Carvalho and Ribeiro (2016), Sun et al. (2017),

Bikker and Vervliet (2018), Mota et al. (2019), Paroush and Schreiber (2019)

and Wang (2023), that a bank’s proﬁtability is directly related to the quality

of its loan assets. Thus, a lower quality of the loan portfolio negatively affects

bank proﬁtability with the amount of provisioning for expected credit losses. It

should also be noted that these authors concluded that this variable is one of

those which most inﬂuences bank proﬁtability. Thus, on the assumption that a

high default rate decreases bank proﬁtability, the following hypothesis is tested:

H1: The sluggishness of banks, as measured by the default rate on the loan

portfolio, negatively affects banks proﬁtability

2.3. LIQUIDITY

Since liquidity is related to a banking institution’s ability to honor its

commitments in relation to depositors’ capital, the level of liquidity of each

148 Marco Amaral

banking institution can be expected to have an impact on banks’ proﬁtability.

The recent Basel III regulatory changes (BCBS, 2013; 2014), introduced two

minimum ﬁnancial liquidity standards such as the liquidity coverage ratio

(LCR) and the net stable funding ratio (NSFR) which, despite being the best

options for measuring bank liquidity, were not used in this study as not all the

ﬁnancial data was available for the period under analysis. Therefore, the ratio

of transformation of customer funds into loans granted was adopted for this

category of explanatory variable. This variable makes it possible to measure

how much customer capital (bank deposits) is used in loans granted to

customers of those same banks. This indicator is widely used by regulators to

measure banks’ liquidity. The higher this indicator, the greater the possibility of

a bank generating income, and therefore becoming more proﬁtable, i.e. banks

with high value for this indicator will be those with a higher degree of leverage,

by converting deposits into loans, thus being appropriate to refer to as factors

of greater bank proﬁtability. The studies carried out by various authors on

this variable are very ambiguous. In fact, while some studies have concluded

a positive relationship between bank proﬁtability and liquidity (Molyneux and

Thornton, 1992; Barth et al., 2013; Trujillo-Ponce, 2013), other studies have

concluded a negative relationship (Bourke, 1989; Demirguc-Kunt and Huizinga,

1999; Hamdi and Hakimi, 2019). Thus, given the mixed character with

different types of positive and negative association, the following hypothesis

was developed to be tested:

H2: Bank liquidity has a positive (negative) effect on banks’ proﬁtability.

2.4. SOLVENCY

With regard to the solvency of banks, this study highlights the importance

of capital ratios, speciﬁcally the Tier 1 solvency ratio, which reﬂects, in

accordance with Basel III guidelines, a bank’s obligation to maintain certain

amounts of capital, known as core capital, to cope with unexpected losses. This

ratio is measured by the relationship between Tier 1 capital and the bank’s

risk-weighted assets (RWA). However, given the difﬁculty in obtaining the data

provided by the banks in the sample, the ﬁnancial data obtained from the

balance sheet resulting from the ratio between equity and total net assets

(solvency understood in accounting terms) was used as an indicator. The higher

this ratio the better, demonstrating, on the one hand, a greater capacity for

banks to cover losses on their assets and, on the other hand, less need to raise

external funds, and therefore greater bank proﬁtability. However, some studies

do not agree on the relationship between solvency and bank proﬁtability.

Bourke (1989), Hoffmann (2011) and Carvalho and Ribeiro (2016) concluded

a negative relationship, but the authors’ studies (Demirguc-Kunt and Huizinga,

1999; Athanasoglou et al., 2008; Kosmidou, 2008; Trujillo-Ponce, 2013; Bikker

and Vervliet, 2018 and Paroush and Schreiber, 2019) point in the opposite

direction, conﬁrming the existence of a positive association. Given that there

149

DETERMINANTS OF THE PROFITABILITY OF SAVING S BANKS IN THE US AND THE MODERATING EFFECT OF THE PANDEMIC CRISIS

REVISTA DE ECONOMÍA MUNDIAL 68, 2024, 141-166

is no agreement on the relationship between the solvency and proﬁtability of

banks, it was pertinent to formulate the following hypothesis to test:

H3: Bank solvency has a positive (negative) effect on bank proﬁtability.

2.5. PRODUCTIVITY

The level of proﬁtability obtained by a banking institution can be inﬂuenced

by its productivity. The productivity variable in this study is translated by the

management indicator called productivity of complementary activity per

employee. In general, this indicator is intended to reﬂect the relationship

between the net commissions generated by each employee of a bank, making

it possible to establish a comparative evaluation measure of the banking

sector. Authors such as Gambacorta and Ibanez (2011) and Trenca et al.

(2015), consider that the fact that an employee generates a high performance

in banking activity will reﬂect in higher bank proﬁtability. This leads to the

following hypothesis to be tested:

H4: Bank productivity has a positive effect on bank proﬁtability.

2.6. MODERATING EFFECT OF THE PANDEMIC CRISIS

The aim of this variable is to analyze how the moderating effect of the

pandemic crisis has inﬂuenced banks’ default, liquidity, solvency and

productivity and what effect it may have had on their proﬁtability. To this

end, a dummy variable was used for the periods without a pandemic crisis (0

corresponds to the periods from 2012 to 2019 and 2022) and for the periods

with a pandemic crisis (1 corresponds to the periods from 2020 and 2021).

Studies such as Wu and Shen (2013) and Sanchez et al. (2017) analyzed the

effects before and after the ﬁnancial crisis and found mixed effects (positive

and negative) between the explanatory variables and bank proﬁtability. In this

sense, the following research hypothesis is proposed:

H5: The moderating effect of the pandemic crisis has a signiﬁcant inﬂuence on

the relationship between the four factors and bank proﬁtability.

3. METHODOLOGY

This section presents the analysis and description of the data, as well as

the sample considered, followed by the variables included in the model and

the treatment of the main statistical data, and concludes with the speciﬁcation

of the econometric model.

3.1. SAMPLE AND DATA DESCRIPTION

The sample consists of a portfolio of 57 ﬁnancial entities operating in the

US banking system in the specialized savings segment (see Appendix I), for

the period from 2012 to 2022, totaling 627 observations. The total volume of

assets and bank deposits of the entities considered in the sample amounts to

150 Marco Amaral

921 billion USD in 2022 (representing 71.9% of total assets, which amount to

1,281 billion USD) and 803 billion USD (representing 74.2% of total deposits,

which amount to 1,082 billion USD), respectively.

The data was collected using the database provided by BankFocus and

Bureau van Dijk, as well as statistical data published by the Federal Deposit

Insurance Corporation (FDIC). Macroeconomic data for the years in the sample

was obtained from the CountryEconomy database of Alldatanow, S.L.

In order to demonstrate the evolution and size of the US banking system

for the savings bank segment, a characterization of the system is given in

Table 3.

TABLE 3. CHARACTERIZATION OF THE SAMPLE: EVOLUTION OF SAVINGS BANKS IN THE US

Description Measure 2012 2019 2020 2021 2022

No. of Institutions Unit 1,011 659 627 607 579

No. of employees Unit 148,918 121,746 122,265 119,610 108,648

Total Assets 1,063 1,154 1,378 1,519 1,281

Of which:

Earning Assets:

- Total Loans

- Other (*)

Non-remunerated Assets

Billion USD

983

652

331

1,096

655

441

1,316

660

656

1,458

681

777

1,215

641

574

Total Deposits

% of Assets

Billion USD

805

75.7

921

79.8

1,139

82.6

1,287

84.7

1,082

84.5

Deposits-to-Loans

(Transformation Ratio) % 81.0 71.1 57.6 52.9 59.2

Equity Capital

% of Assets

Billion USD

127

11.9

125

10.8

134

9.7

147

9.7

7.3

Net Interest Margin

Of which:

Yield on Earnings Assets

Cost of Funding Earning Assets

3.46

4.38

0.92

3.87

4.83

0.96

3.09

3.65

0.56

2.78

3.06

0.28

3.28

3.73

0.45

Income

Of which:

Interest Income

Non-interest Income

Billion USD

Provisions for Credit Losses Billion USD 5 5 8 0,3 4

Net Income Billion USD 11 15 11 16 13

(*) Other Earnings Assets: Securities, Derivatives, Depository Institutions and other ﬁnancial

instruments.

Source: Author´s own creation based on publications of FDIC statistical data.

Table 3 shows that in 2022 the FDIC supervised 579 savings banks with

108,648 employees and total assets and deposits of 1,281 billion US dollars

and 1,082 billion US dollars, respectively. Although during the sample period

total assets grew by more than 20% (from 1,063 in 2012 to 1,281 in 2022)

and total deposits by more than 30% (from 805 in 2012 to 1,082 in 2022),

the same was not true of the number of institutions and employees, with a

sharp reduction of 432 savings institutions (-42.7%) and 40,270 employees

151

DETERMINANTS OF THE PROFITABILITY OF SAVING S BANKS IN THE US AND THE MODERATING EFFECT OF THE PANDEMIC CRISIS

REVISTA DE ECONOMÍA MUNDIAL 68, 2024, 141-166

(-27.0%). In turn, the net interest margin rate, which corresponds to the

ratio between net interest income and the average balance of total interest-

generating assets, has remained stable throughout the analysis period (above

3%), with the exception of the year 2021, when there was a rate of 2.78%,

justiﬁed by a lower effect of household consumption in the period of the

pandemic crisis and greater reinforcement in terms of savings. It should also

be noted that there has been a change in the evolution of earning assets,

which grew by 23.6% (from 983 in 2012 to 1,215 in 2022). It should be

noted that customer loans remain stable in their composition, while there

has been a sharp increase in other earning assets such as debt securities

and other instruments, which grew from 331 billion US dollars in 2012 to

574 billion US dollars in 2022 (an increase of 73.4%), having reached the

highest amount in 2021 of 777 billion US dollars, an amount that for the

ﬁrst time was higher than loans to customers, which amounted to 681 billion

US dollars. It can be concluded that, given the change in the type of earning

assets, income has remained stable (above 50 billion dollars), although other

income (such as commissions and others) has fallen substantially (-52.4%)

since 2012 (21 billion dollars) and will only reach 10 billion dollars in 2022.

However, despite the fact that income fell slightly in the period under review

(in 2012, 54 billion dollars, against 51 billion dollars in 2022), there was

a slight improvement in net income (in 2012, 11 billion dollars, against 13

billion dollars in 2022) as a result of tight and strong management in cost

containment, namely with a sharp reduction in physical branches and human

resources. Lastly, the level of provisioning for credit losses for the year 2021

was signiﬁcantly reduced (0.3 billion dollars) as 82 ﬁnancial institutions

adopted the new methodologies for accounting for current expected credit

losses (CECL), as published in the FDIC’s quarterly information report (FDIC,

2022).

3.2. STUDY VARIABLES

The dependent variable of bank proﬁtability is measured in this study by

two alternative indicators of bank performance, RoA and RoE, according to

the authors’ studies (Athanasoglou et al., 2008; Tregenna, 2009; Sanchez et

al., 2017; Bikker and Vervliet, 2018 and Veeramoothoo and Hammoudeh,

2022). From the literature reviewed in this paper, it was felt that these are

the indicators most commonly adopted by authors who have addressed this

issue. They are also indicators widely used by regulators and banks to assess

a bank’s performance. While return on equity (RoE) measures the relationship

between net income and equity, allowing us to assess the performance of the

resources placed by investors in the bank, return on assets (RoA) measures the

relationship between net income and assets. This ratio assesses the proﬁtability

generated by the assets ﬁnanced by the bank.

The recent evolution of these two indicators in the US banking system, for

the savings bank segment supervised by the FDIC, was as follows (see Graph 1).

152 Marco Amaral

GRAPH 1 – EVOLUTION OF THE DEPENDENT VARIABLES IN THE US BANKING SYSTEM

Source: Obtained using STATA software.

Over the last 11 years, the North American banking sector in the savings

bank segment has shown a trend towards improved bank proﬁtability, both in

terms of RoE (more pronounced) and RoA (more stable). However, it can be

seen that the number of ﬁnancial institutions (numbmean) over this period has

decreased signiﬁcantly, as opposed to the volume of assets (atmean) which

has increased.

The independent variables used in this study are reﬂected in the hypotheses

previously formulated and are indicators and ratios related to the activity of the

banking sector and represent a set of variables that identify ﬁve hypotheses to

be tested. Thus, the variables included in the model, as well as how they are

determined and the source of the data, are as follows (see Table 4).

TABLE 4. VARIABLES INCLUDED IN THE MODEL

Variables Notation Form of Determination Source

Dependent:

Return on Assets RoA RoA ratio (Return on Assets) =

Net Proﬁt / Net Assets BankFocus

Return on Equity RoE RoE ratio (Return on Equity) =

Net Proﬁt / Shareholders’ Equity BankFocus

(Continue)

153

DETERMINANTS OF THE PROFITABILITY OF SAVING S BANKS IN THE US AND THE MODERATING EFFECT OF THE PANDEMIC CRISIS

REVISTA DE ECONOMÍA MUNDIAL 68, 2024, 141-166

Independent:

Default DEF

Customer Loan and Advances Default Rate =

Loan Impairment (year) / Net Loans and Advances

to Customers

BankFocus

Liquidity LIQ Transformation Ratio =

Net Customer Loans / Customer Deposits BankFocus

Solvency SOL Solvency Ratio (in accounting terms) =

Equity / Net Assets BankFocus

Productivity PRO Degree of Production of Complementary Activity =

Net Commissions / No. of Employees BankFocus

Moderation:

Pandemic Crisis CRI

Dummy variable (binary) where:

0 - Without Pandemic Crisis (periods from 2012 to

2019 and 2022);

1 - With Pandemic Crisis (2020 and 2021)

Author

Control:

Macroeconomic:

Inﬂation INF

Consumer Price Index

(%)

Country Eco-

nomy

Central Bank

Reference Rate CBR US Central Bank Interest Rate

(%)

Country Eco-

nomy

Type of Govern-

ment TGO

Dummy variable (binary) where:

0 - Democratic Government;

1 - Republican Government

Author

Banks:

Dimension DIM Logarithm (natural) =

of Net Asset Value BankFocus

Source: Author´s own creation.

3.3. ECONOMETRIC MODEL AND STATISTICAL METHODS

The studies that evaluate the factors that determine the proﬁtability of

banking institutions, such as Trujillo-Ponce (2013), Serrano and Pavia (2014),

Carvalho and Ribeiro (2016), Bikker and Vervliet (2018), Hamdi and Hakimi

(2019), Mota et al. (2019), Paroush and Schreiber (2019), Mendoza et al.

(2020), Otero et al. (2021), Veeramoothoo and Hammoudeh (2022) and

Wang (2023) basically use linear regression models, which make it possible

to describe and assess which independent variables have explanatory power

over the dependent variables. Thus, the proposed model, including the control

variables for the study of the determinants of proﬁtability and their moderating

effects of the pandemic crisis on bank performance can be illustrated in the

following Figure 1.

The data in this study includes the descriptive statistics of the study

variables, as well as Pearson’s correlation analysis between the variables for

the period analyzed between 2012 and 2022 (11 years), as shown in Table 5.

TABLE 5. DESCRIPTIVE STATISTICS AND CORRELATION ANALYSIS

RoA RoE DEF LIQ SOL PRO INF CBR DIM TGO

(a)

CRI

(a)

Mean 1.08 9.20 0.19 85.40 12.60 19.35 2.55 0.98 15.37 - -

S. D. 1.09 6.18 0.49 23.59 9.45 32.13 2.04 1.24 1.18 - -

Min. -9.68 -12.41 -1.50 2.52 2.39 -57.68 0.70 0 10.07 - -

Max. 13.03 45.53 4.67 192.6 96.94 306.23 7.00 4.25 18.60 - -

(Continue)

154 Marco Amaral

No.

Obs. 627 627 620 620 627 616 627 627 627 627 627

RoA 1

RoE 0.73 1

DEF -0.03 -0.03 1

LIQ 0.27 -0.01 -0.05 1

SOL 0.58 -0.04 -0.05 0.37 1

PRO -0.21 -0.10 -0.03 -0.24 -0.17 1

INF 0.10 0.19 -0.12 -0.13 -0.09 0.12 1

CBR 0.02 0.15 -0.05 -0.01 -0.09 0.10 0.46 1

DIM -0.10 0.01 0.10 0.07 -0.17 0.21 0.23 0.16 1

TGO -0.01 -0.03 0.11 0.04 0.02 -0.05 -0.23 0.16 0.10 1

CRI 0.05 0.03 0.09 -0.12 -0.02 0.01 0.38 -0.37 0.17 0.13 1

(a) Dummy variable.

Source: Author´s own creation.

FIGURE 1 – PROPOSED MODEL

Source: Author´s own creation.

It can be seen that bank proﬁtability, measured by RoA and RoE, for the

period under analysis, averaged 1.08% and 9.20%, respectively. On the other

hand, the correlation coefﬁcients are generally not signiﬁcantly high (below or

above 75%) to cause concern about multicollinearity problems.

In order to answer the ﬁve hypotheses mentioned above, the following four

multiple linear regression models were developed, as shown in Table 6.

TABLE 6. LINEAR REGRESSION MODELS

RoAit = β0 + β1FATit + β2CRIit + β3VCit + εit Model 1

RoAit = β0 + β1FATit + β2CRIit + β3CRIit * INTFATit + β4VCit + εit Model 2

RoEit = β0 + β1FATit + β2CRIit + β3VCit + εit Model 3

RoEit = β0 + β1FATit + β2CRIit + β3CRIit * INTFATit + β4VCit + εit Model 4

Where,

(Continue)

155

DETERMINANTS OF THE PROFITABILITY OF SAVING S BANKS IN THE US AND THE MODERATING EFFECT OF THE PANDEMIC CRISIS

REVISTA DE ECONOMÍA MUNDIAL 68, 2024, 141-166

RoAit represents the return on assets ratio of institution i at time t;

RoEit represents the return on equity ratio of institution i at time t;

FATit represents the determining factors of institution i at time t;

In which there are four determining factors:

DEFit represents the default rate on loans to customers of institution i at time t;

LIQit represents the ratio of deposits to loans of institution i at time t;

SOLit represents the solvency ratio of institution i at time t;

PROit represents the production of complementary activity of institution i at time t;

CRIit represents the dummy of pandemic crisis of institution i at time t;

VCit represents the control variables of institution i at time t;

They are composed of four control variables:

INFit represents the inﬂation rate for institution i at time t;

CBRit represents the central bank reference rate for institution i at time t;

TGOit represents the type of government for institution i at time t;

DIMit represents the logarithm of the size of net assets institution i at time t;

CRISit * INTFATit represents the moderating effect of the interaction between the crisis and the determining fac-

tors of institution i at time t;

In which they are constructed by four crossed variables (see Appendix II):

DEFit*CRIit represents the interaction between default and the crisis of institution i at time t;

LIQit*CRIit represents the interaction between liquidity and the crisis of institution i at time t;

SOLit*CRIit represents the interaction between solvency and the crisis of institution i at time t;

PROit*CRIit represents the interaction between productivity and crisis of institution i at time t;

β0 is the constant term;

εit is the statistical error term of institution i at time t.

Source: Author´s own creation.

3.4. RESULTS AND DISCUSSION

The econometric estimation of the model uses the Panel Data technique

(Stata - Statistics Data Analysis) which combines cross-section data (banks)

and time-series data (years), obtaining a strongly balanced panel.

In order to model the functional relationship between the variables,

multiple linear regression models were used, using a panel data model,

and the Hausman test was applied in order to assess whether this method

best ﬁts the ﬁxed-effects Ordinary Least Square (OLS regression) model or

the random-effects Generalized Least Squares (GLS regression) model. The

analysis showed that the model with the highest quality is the random effects

model. According to Wooldridge (2002), the random effects model is more

efﬁcient in large samples because the random effects estimators have smaller

standard errors.

3.4.1. ESTIMATION OF THE ECONOMETRIC RESULTS

In order to estimate the model, it is necessary to take into account that the

data treatments used in this study are arranged in a longitudinal panel, made

up of a set of different entities (N=57 banks) and over various periods of time

(T=11 years). Therefore, in order to model the functional relationship between

the variables, four multivariate linear regression models were adopted, using

the generalized least squares random effects model, as shown in Table 7.

156 Marco Amaral

TABLE 7. REGRESSION RESULTS

Variable RoA (Return on Assets) RoE (Return on Equity)

RE (Random Effects)

Model 1 Model 2 Model 3 Model 4

DEF -0.0866* -0.0128 -0.7276* -0.3770

LIQ 0.0064*** 0.0063*** 0.0535*** 0.0450***

SOL 0.0620*** 0.0590*** -0.4484*** -0.4536***

PRO -0.0040*** -0.0045*** -0.0321*** -0.0332***

CRI 0.0152 -0.9745*** -0.1735 -9.5107***

INF 0.0616*** 0.0434** 0.3775** 0.3000*

CBR -0.0080 0.0059 0.2116 0.2237

TGO 0.0099 -0.0051 -0.2990 -0.3628

DIM 0.0310 0.0646 0.8993** 1.1974***

DEF * CRI -0.2357*** -1.2861*

LIQ * CRI 0.0052** 0.0701***

SOL * CRI 0.0569*** 0.3558***

PRO * CRI 0.0010 -0.0003

_CONS -0.7226 -1.1663 -4.0365 -7.6483

No. Observations: 609 609 609 609

No. Banks: 57 57 57 57

R-sq.

Within 0.1635 0.2241 0.1837 0.2393

Between 0.3937 0.3696 0.0087 0.0140

Overall 0.3338 0.3435 0.0166 0.0253

Rho 0.5824 0.5896 0.5731 0.5776

Hausman:

chi²1.55 13.70 29.17 39.41

Prob> chi²0.9968 0.3954 0.0600 0.0520

(*) (**) (***) Statistically signiﬁcant results for a signiﬁcance level of 0.10, 0.05 and 0.01 respectively.

Source: Author´s own creation.

Table 7 shows the results obtained when estimating the regression model,

which was run four times, based on the proﬁtability of US banks specializing in

the savings segment, measured by the RoA (models 1 and 2) and RoE (models

3 and 4). While models 1 and 3 allow us to assess the relationship between

bank proﬁtability and the explanatory factors, models 2 and 4 allow us to

assess the moderating effects of the pandemic crisis on the inﬂuence of the

relationship between the factors and bank proﬁtability. Thus, the results indicate

that there is an acceptable level of explanation of the determining factors for

the RoA indicators with an adjusted R² of 33.38% and 34.35%. Regarding

the RoE indicator, there is a low adjusted R² of 1.66% and 2.53%. As far as

the explanatory variables are concerned, the results obtained show statistical

signiﬁcance in explaining bank proﬁtability as measured by RoA and RoE for

the four determining factors under study, i.e. the variables of default (DEF),

liquidity (LIQ), solvency (SOL) and productivity (PRO). With regard to the control

variables, inﬂation (INF) and bank size (DIM) also show statistical signiﬁcance.

157

DETERMINANTS OF THE PROFITABILITY OF SAVING S BANKS IN THE US AND THE MODERATING EFFECT OF THE PANDEMIC CRISIS

REVISTA DE ECONOMÍA MUNDIAL 68, 2024, 141-166

Finally, it can be seen that the moderating effect of the pandemic crisis, the

variables of default (DEF), liquidity (LIQ) and solvency (SOL) were the variables

that had an inﬂuence on both bank proﬁtability indicators (RoA and RoE). It

should also be noted that a complementary analysis to this study was carried

out to highlight the sensitivity of the linear regression estimates (see Appendix

III). The data was removed for the periods considered during the pandemic crisis

(2020 and 2021) and the results show that the default (DEF) variable was not

statistically signiﬁcant. We then analyzed the moderating effect of the interaction

between the pandemic crisis and default on bank proﬁtability and found that

when the years of the pandemic crisis were included in models 1 and 3, it showed

a negative sign and became statistically signiﬁcant for both indicators (RoA and

RoE) at a 10% signiﬁcance level. Thus, the results show how the crisis negatively

moderated the effect of default on the proﬁtability of US banks.

3.4.2. DISCUSSION OF RESULTS

This section ﬁrst presents the results of the main effects of the dependent

variables testing hypotheses 1, 2, 3 and 4. Next, the moderating effects

related to hypothesis 5 are presented. Thus, the results of the hypotheses to

be tested are as follows:

H1: The sluggishness of banks, as measured by the default rate on the loan

portfolio, negatively affects banks’ proﬁtability.

The results show that the explanatory variable in the model relating to

the default rate (DEF) contributes to the decrease in proﬁtability of savings

banks in the USA. The regression coefﬁcients for the RoA and RoE variables

in the four models show a negative sign, as expected, and are statistically

signiﬁcant for models 1 and 3. In this case, they show a statistical signiﬁcance

level of 10% for the RoA and RoE indicator. It can therefore be said that when

banks’ customer defaults increase, their proﬁtability decreases. These results

are in line with the empirical evidence of Demirguç-Kunt and Huizinga (1999),

Athanasoglou et al. (2008), Bikker and Vervliet (2018), Paroush and Schreiber

(2019) and Wang (2023). In this way, hypothesis 1 can be validated.

H2: Bank liquidity has a positive (negative) effect on banks’ proﬁtability.

Bank proﬁtability varies with bank liquidity, and the results show that

the independent variable of liquidity (LIQ) has a positive relationship with

the proﬁtability of US savings banks. For the four models tested, there is a

statistical signiﬁcance level of 1% for both proﬁtability indicators (RoA and

RoE). Although a mixed association between the variables was expected, it is

nevertheless possible to validate hypothesis 2, since it is possible to state that

banking institutions can obtain greater bank proﬁtability by increasing their

liquidity indicators. These results are in line with studies by Molyneux and

Thornton (1992), Barth et al. (2013) and Trujillo-Ponce (2013).

H3: Bank solvency has a positive (negative) effect on bank proﬁtability.

In the case of proﬁtability measured by RoA, the sign of the regression

coefﬁcient of the two models is positive and statistically signiﬁcant, with a

158 Marco Amaral

statistical signiﬁcance level of 1%, so an increase in the solvency indicator

(SOL) contributes to an increase in bank proﬁtability. In contrast, proﬁtability as

measured by RoE shows a negative and statistically signiﬁcant relationship for

both models at a 1% signiﬁcance level. The results show a positive relationship

for the RoA variable and a negative relationship for the RoE variable, as

expected. Hypothesis 3 can therefore be validated, since a mixed relationship

was predicted between the variables under study. This result is in line with the

studies by Bourke (1989) Hoffmann (2011) and Carvalho and Ribeiro (2016),

who concluded a negative relationship, and with the studies by Demirguc-Kunt

and Huizinga (1999), Athanasoglou et al. (2008), Kosmidou (2008), Trujillo-

Ponce (2013), Bikker and Vervliet, (2018) and Paroush and Schreiber (2019),

who concluded a positive relationship.

H4: Bank productivity has a positive effect on bank proﬁtability.

The estimated results show that the productivity (PRO) of US savings banks

in terms of the ratio of complementary activity output, measured by the ratio

of commission generated by each employee, has a signiﬁcant inﬂuence on

bank proﬁtability. The results obtained for the four estimated models show

a statistically signiﬁcant coefﬁcient of 1%, which validates hypothesis 4.

However, the coefﬁcients obtained for this indicator were negative, contrary to

the expected sign, which allows us to infer that the proﬁtability of RoA and RoE

decreased. This result disagrees with the empirical evidence of Gambacorta

and Ibanez (2011) and Trenca et al. (2015), who concluded that there is a

positive association between productivity and bank proﬁtability. Therefore,

this could be a ﬁeld for further study of the results obtained and a considerable

expansion of research on this subject.

H5: The moderating effect of the pandemic crisis has a signiﬁcant inﬂuence on

the relationship between the four factors and bank proﬁtability.

The moderating effect of the pandemic crisis has a statistically signiﬁcant

inﬂuence on the relationship between three of the four determining factors

and on the RoA and RoE proﬁtability measures (Models 2 and 4). According

to Hayes (2013), an interaction is the product of the relationship between

each of the determining factors and the moderating effect. Thus, by analyzing

the interaction of the product of each of the variables under study (DEF*RIS;

LIQ*RIS; SOL*CRI; PRO*CRI) with the moderating effect of the pandemic

crisis, it can be concluded that the pandemic crisis has a moderating effect on

the relationship between the determining factors and proﬁtability. In turn, this

effect mainly inﬂuences the variables of default (DEF*CRI), liquidity (LIQ*CRI)

and solvency (SOL*CRI), with a greater magnitude in default (DEF*CRI), since

it ﬂows negatively and has an impact on the proﬁtability indicator measured by

RoA (Coef = -0.2357) and the proﬁtability indicator measured by RoE (Coef =

-1.2861). Thus, through the moderating effect, we can see how much the effect

of the default variable (DEF*CRI) on the proﬁtability variables (RoA and RoE)

varies when we are in a period of crisis. As the default variable (DEF*CRI) shows

a level of statistical signiﬁcance of 1% and 10% for RoA and RoE, respectively,

it can be concluded that the crisis negatively moderated the effect of default

159

DETERMINANTS OF THE PROFITABILITY OF SAVING S BANKS IN THE US AND THE MODERATING EFFECT OF THE PANDEMIC CRISIS

REVISTA DE ECONOMÍA MUNDIAL 68, 2024, 141-166

(DEF*CRI) on proﬁtability as measured by both RoA and RoE, i.e. banks with

higher default rates during the pandemic crisis had greater difﬁculties in their

proﬁtability. On the other hand, the results show that the crisis positively

moderated the effect of solvency (SOL*CRI) and liquidity (LIQ*CRI) on

proﬁtability. This evidence suggests that US savings banks with better solvency

and liquidity had fewer difﬁculties during the pandemic crisis. In turn, the effect

of the crisis is null for the productivity variable (PRO*CRI) as it does not have

a signiﬁcant inﬂuence. These results are in line with the studies by Wu and

Shen (2013) and Sanchez et al. (2017), who found mixed effects (positive and

negative) between the explanatory variables and bank proﬁtability.

4. CONCLUSION AND LIMITATIONS OF THE STUDY

Although there has been a sharp decline in the number of savings institutions

in the United States in recent years, there continues to be an upward trend in

bank proﬁtability.

This paper aims to analyze the impact of some factors speciﬁc to the

banking sector on the ﬁnancial proﬁtability of US savings banks for the period

from 2012 to 2022 and to assess the moderating effect that the pandemic

crisis has had on this sector. To this end, four econometric models were built,

which explain and can be used to predict the determinants of bank proﬁtability

and the moderating role of the pandemic in these effects.

Thus, several conclusions emerge from this study. Thus, for a total of 609

observations corresponding to 57 savings banks, it can be seen that the banks

in the sample did not obtain economic beneﬁts in their ﬁnancial performance

in the four explanatory factors. The liquidity factor (LIQ) and the solvency

factor (SOL) have a clear positive effect on bank proﬁtability, with a statistical

signiﬁcance level of 1%. These results mean that a higher degree of leverage,

through the transformation of customer deposits into loans and advances to

customers (LIQ), together with robust equity capital (SOL), can translate into

higher proﬁtability for banks. On the other hand, the default factor (DEF) shows

a negative relationship at a signiﬁcance level of 10% for RoA and RoE, and it

can be inferred that poorer credit portfolio quality negatively inﬂuences bank

proﬁtability. Finally, the results also emphasize the productivity factor (PRO),

showing, contrary to the expected sign, a negative association with bank

proﬁtability at a statistical signiﬁcance level of 1%. This unexpected result in

terms of revenue generation per employee could be directly related to the

sharp loss of commission income by North American banks in the period under

analysis, which, together with the sharp reduction in the number of employees,

as shown in Table 3, could translate into lower bank proﬁtability. Therefore, this

could be a ﬁeld for further research into, for example, different business models

that could lead to a generation of revenue per employee in future research.

Regarding the moderating effect of the pandemic crisis, the results show that

the crisis positively moderated the effect of solvency (SOL * CRI) and liquidity

(LIQ * CRI) on bank proﬁtability. Thus, banks with better solvency and liquidity

160 Marco Amaral

had fewer difﬁculties during the pandemic period. In contrast, banks with high

default rates during the crisis had greater difﬁculties in their proﬁtability, i.e.

the crisis strongly negatively moderated the effect of default (DEF * CRI), which

was the variable with the greatest magnitude on bank proﬁtability in the USA.

Finally, the interaction effect of the pandemic crisis on the relationship between

productivity (PRO * CRI) and proﬁtability is insigniﬁcant.

It is therefore felt that banks should have a contingency plan for the

occurrence of similar situations, namely a “buffer” of contingent capital to

temporarily strengthen the robustness of banks and that the role of banking

regulators and supervisors could also be more proactive in implementing

medium and long-term ﬁnancial instruments to restore proﬁtability.

According to the individual analysis of the p> |z| test, the control variables

of the central bank reference rate (CBR) and the type of government in the

US (TGO) do not have an impact on bank proﬁtability. However, the bank size

variable (DIM) has a positive effect on bank proﬁtability as measured by the

RoE indicator at a statistical signiﬁcance level of 1% and 5%, i.e. an increase

in RoE implies an increase in proﬁtability. In turn, the inﬂation (INF) variable

positively inﬂuences bank proﬁtability for the four models tested, i.e. when the

inﬂation rate increases, bank proﬁtability also increases.

The results of this study can be useful for bank decision-makers, as well

as supervisors and regulators in the banking sector, as they identify valuable

guidance on the effect of key ﬁnancial variables on a bank’s proﬁtability. It

could also be used to develop a framework for policies and regulations that

affect bank proﬁtability.

However, the study has some limitations, since the sample used is small,

limited to ﬁfty-seven banks in the North American banking system for the

savings segment. In addition, the period of analysis is medium-term (11

years) and some ﬁnancial indicators, due to the lack of ﬁnancial data, require

regulatory support, so the results obtained should be analyzed taking this

limitation into account.

Finally, it is suggested that future research should extend the sample and

the period of analysis and consider other indicators, whether of a ﬁnancial

nature (tier 1 solvency ratio, liquidity coverage ratio - LCR, net stable funding

ratio - NSFR, technology systems, digital transition, environment, corporate

governance or business models) or of a non-ﬁnancial nature (GDP, active

and passive market interest rates or the unemployment rate), which could

potentially explain bank proﬁtability.

ACKNOWLEDGEMENTS

The author is grateful to the anonymous referees of the journal for their

extremely useful suggestions to improve the quality of the article.

161

DETERMINANTS OF THE PROFITABILITY OF SAVING S BANKS IN THE US AND THE MODERATING EFFECT OF THE PANDEMIC CRISIS

REVISTA DE ECONOMÍA MUNDIAL 68, 2024, 141-166

REFERENCES

Alshatti, A. S. (2016). Determinants of bank´s proﬁtability – The case of Jordan.

Investment Management and Financial Innovations, 13(1), p. 84-91.

Athanasoglou, P. P.; Brissimis, S. N., Delis, M. D. (2008). Bank-speciﬁc, industry-

specifc and macroeconomic determinants of bank proﬁtability. Journal of

International Financial Markets, Institutions and Money, v. 18, n.º 2, p.

121-136.

Barata, J. M., (2014). Determinantes da Rendibilidade Bancária – Um estudo

empírico. Revista Inforbanca nº 102, outubro-dezembro, p. 26-28.

Barth, J.; Lin, C.; Ma, Y.; Seade, J.; Song, F. M. (2013). Do bank regulation,

supervision and monitoring enhance or impede bank efﬁciency? Journal of

Banking and Finance, 37, p. 2879-2892.

BCBS (Basel Committee on Banking Supervision), (2013). “Basel III: The liquidity

coverage ratio and liquidity risk monitoring”, BIS – Bank for International

Settlements, january.

BCBS (Basel Committee on Banking Supervision), (2014). “Basel III: The net

stable funding ratio”, BIS – Bank for International Settlements, october.

Bikker, J. A.; Vervliet, T. M. (2018). Bank proﬁtability and risk-taking under low

interest rates. International Journal of Finance and Economics, 23, 3-18.

Bourke, P. (1989). Concentration and other determinants of bank proﬁtability

in Europe, North America and Australia. Journal of Banking and Finance,

13(1), p. 65-79.

Carvalho, J. F.; Ribeiro, A. (2016). Fatores explicativos da rendibilidade do

setor bancário: evidência empírica em Portugal. Estudos do ISCA, série IV,

n.º 13, p. 1-11.

Demirguç-Kunt, A.; Huizinga, H. (1999). Determinants of commercial bank

interest margins and proﬁtability: some international evidence. World Bank

Economic Review, 13(2), p. 379-408.

DeYoung, R.; Rice, T. (2004). Non interest income and ﬁnancial performance at

US commercial banks. The Financial Review, 39, p. 101-127.

Duraj, B.; Moci, E. (2015). Factors inﬂuencing the bank proﬁtability – Empirical

evidence from Albania. Asian Economic and Financial Review, 5(3), p. 483-

494.

Elisa, M.; Guido, P. (2016). Factors affecting bank proﬁtability in Europe: An

empirical investigation. African Journal of Business Management, 10(17),

p. 410-420.

FDIC (Federal Deposit Insurance Corporation), (2022). “FDIC Quarterly”, 16(1),

p. 1-54.

Fronk, J. (2016). Core proﬁtability of community banks: 1985-2015. FDIC

Quarterly, 10(4), p. 37-46.

Gambacorta, L.; Ibanez, M. D. (2011). The bank lending channel: lessons from

the crisis. Economic Policy, April, p. 135-182.

162 Marco Amaral

Hamdi, H.; Hakimi, A. (2019). Does liquidity matter on bank proﬁtability?

Evidence from a nonlinear framework for a large sample. Business and

Economics Research Journal, vol. 10, n.º 1, p. 13-26.

Hayes, A. F. (2013). Introduction to mediation, moderation and conditional

process analysis: A regression-based approach. Guilford Press.

Hoffmann, P. (2011). Determinants of the proﬁtability of the US banking

industry. International Journal of Business and Social Science, 2(22), p.

255-269.

Islam, Md. A.; Nazirul, Md. I. S.; Mahahub, R.; Ariﬁn, S.; Shaﬁullah, A.

(2017). Determinants of proﬁtability of commercial bank´s in Bangladesh.

International Journal of Banking and Financial Law, 1(1), p. 1-11.

Kosmidou, K. (2008). The determinants of banks proﬁts in Greece during the

period of EU ﬁnancial integration. Managerial Finance, vol. 34, n.º 3, p.

146-159.

Le, T. D.; Ngo, T. (2020). The determinants of bank proﬁtability: A cross-country

analysis. Central Bank Review, 20(2), p. 65-73.

Liang, H. Y.; Ching, Y. P., Chan, K. C. (2013). Enhancing bank performance

through branches or representative ofﬁces? Evidence from European

banks. International Business Review, 22(3), p. 495-508.

Mendoza, J. A. M.; Yelpo, S. M. S.; Ramos, C. L. V.; Fuentealba, C. D. (2020). Are

the effects of market concentration and income diversiﬁcation on banking

performance persistent? Ecos de Economía; A Latin American Journal of

Applied Economics 24(50), p. 25-44.

Messai, A. S.; Mohamed, I. G.; Fathi, J. (2015). Determinants of bank

proﬁtability in Western European Countries: Evidence form system GMM

estimates. International Business Research, 8(7), p. 30-42.

Molyneux, P.; Thornton, J. (1992). Determinants of European bank proﬁtability:

a note. Journal of Banking and Finance, 16(6), p. 1173-1178.

Mota, C.; Silva, E. S. (2019). Determinantes da rentabilidade bancária:

evidências para os maiores bancos portugueses. European Journal of

Applied Business Management, 5(2), p. 78-96.

Otero, G. L.; Durán-Santomil, P; Alaraj, R. (2021). Corporate Governance and

Mena banks performance. Revista Economía Mundial, nº 58, p. 51-74.

Paroush, J.; Schreiber, B. Z. (2019). Proﬁtability, capital, and risk in US

commercial and savings banks: Re-examination of estimation methods. The

Quarterly Review of Economics and Finance, 74, p. 148-162.

Sanchez, P. E.; Gonzalez, M. C.; Gazquez, J. D. P. (2017). Corporate social

performance and its relation with corporate ﬁnancial performance:

International evidence in the banking industry. Journal of Cleaner

Production, 162, p. 1102-1110.

Serwadda, I. (2018). Determinants of commercial bank´s proﬁtability. Evidence

from Hungary. Acta Universitatis Agriculturae et Silviculturae Mendelianae

Brunensis, 66(5), p. 1325-1335.

Serrano, S. C.; Pavia, J. M. (2014). Determinantes y diferencias en la rentabilidad

de cajás y bancos. Revista de Economía Aplicada, 22, p. 117-154.

163

DETERMINANTS OF THE PROFITABILITY OF SAVING S BANKS IN THE US AND THE MODERATING EFFECT OF THE PANDEMIC CRISIS

REVISTA DE ECONOMÍA MUNDIAL 68, 2024, 141-166

Sun, P. H.; Mohamad, S.; Ariff, M. (2017). Determinants driving bank

performance: a comparison of two types of banks in the OIC. Paciﬁc-Basin

Financial Journal, 42, p. 193-203.

Titko, J.; Viktorija, S.; Daiva, J. (2015). Drivers of bank proﬁtability: Case of

Latvia and Lithuania. Intellectual Economics, 9(2), p. 120-129.

Tregenna, F. (2009). The fat years: The structure and proﬁtability of the US

banking sector in the pre-crisis period. Cambridge Journal of Economics,

33, p. 609-632.

Trenca, I.; Petria, N.; Corovei, E. A. (2015). Impact of macroeconomic variables

upon the banking system liquidity. Procedia Economics and Finance, 32, p.

1170-1177.

Trujillo-Ponce, A. (2013). What determines the proﬁtability of banks? Evidence

from Spain. Accounting and Finance, 53(2), p. 561-586.

Veeramoothoo, S.; Hammoudeh, S. (2022). Impact of Basel III liquidity

regulations on U.S. Bank performance in different conditional proﬁtability

spectrums. North American Journal of Economics and Finance, 63, p. 1-18.

Wang, L. (2023). Central bank asset purchases, bank´s risk security holdings

and proﬁtability: Macro and micro evidence from Japan and the U.S..

International Review of Economics and Finance, 87, p. 347-364.

Wahdan, M.; Leithy, W. E. (2017). Factor affecting the proﬁtability of commercial

banks in Egypt over the last 5 year (2011-2015). International Business

Management, 11(2), p. 342-349.

Wu, M. W.; Shen, C.H. (2013). Corporate social responsibility un the banking

industry: Motives and ﬁnancial performance. Journal Banking and Finance,

33(9), p. 3529-3547.

Wooldridge, J. M. (2002). Econometric Analysis of Cross Section and Panel

Data. The MIT Press, Cambridge, Massachusetts, London, England.

164 Marco Amaral

APPENDIX I | COMPLETE LIST OF US SAVINGS BANKS CONSIDERED IN THE SAMPLE AND

NET ASSET VALUE FOR THE YEAR 2022 IN BILLIONS OF USD

Nº Sample Savings Banks Assets %

1 AMERICAN SAVINGS BANK, FSB 9,5 0.7

2 AMERIPRISE BANK, FSB 19,0 1.5

3 ATLANTIC UNION BANK 20,3 1.6

4 AXOS BANK 17,9 1.4

5 BANGOR SAVINGS BANK 7,4 0.6

6 BEAL BANK 6,7 0.5

7 BROOKELINE BANK 6,1 0.5

8 CAMBRIDGE SAVINGS BANK (MHC) 6,5 0.5

9 CAPITOL FEDERAL SAVINGS BANK (MHC) 10,0 0.8

10 COLUMBIA BANK 10,1 0.8

11 DOLLAR BANK, FSB 11,4 0.9

12 EL DORADO SAVINGS BANK, FSB 2,7 0.2

13 FIDELITY BANK, NATIONAL ASSOCIATION 3,1 0.2

14 FIRST FEDERAL BANK 3,8 0.3

15 FIRST FEDERAL SAVINGS & LOAN ASSOCIATION OF LAKEWOOD 2,5 0.2

16 FIRST FOUNDATION BANK 13,0 1.0

17 FIRSTRUST SAVINGS BANK 5,2 0.4

18 FLAGSTAR BANK, NATIONAL ASSOCIATION 90,0 7.0

19 GATE CITY BANK 3,4 0.3

20 HOME BANK, NATIONAL ASSOCIATION 3,2 0.3

21 HOME FEDERAL BANK OF TENNESSEE 2,8 0.2

22 HOMETRUST BANK 3,6 0.3

23 JOHN DEERE FINANCIAL, FSB 3,1 0.2

24 LENDINGCLUB BANK, NATIONAL ASSOCIATION 7,6 0.6

25 LIBERTY BANK 6,9 0.5

26 LUTHER BURBANK SAVINGS 8,0 0.6

27 MASCOMA BANK 2,7 0.2

28 MIDDLESEX SAVINGS BANK 6,2 0.5

29 MIDFIRST BANK 34,7 2.7

30 MORGAN STANLEY PRIVATE BANK NATIONAL ASSOCIATION 119,9 9.4

31 NORTH AMERICAN SAVINGS BANK, FSB 2,5 0.2

32 NORTH SHORE BANK 2,6 0.2

33 NORTHEAST BANK 2,8 0.2

34 NORTHWEST BANK 14,2 1.1

35 NORTHWEST BANK SUBSIDIARES 2,5 0.2

36 OCEANFIRST BANK, NATIONAL ASSOCIATION 13,0 1.0

37 PACIFIC PREMIER BANK 21,7 1.7

38 PATHWARD, NATIONAL ASSOCIATION 6,7 0.5

39 PREMIER BANK 8,4 0.7

40 PRINCIPAL BANK 8,4 0.7

41 PROVIDENT BANK 13,8 1.1

42 RAYMOND JAMES BANK 42,1 3.3

43 RIDGEWOOD SAVINGS BANK 6,8 0.5

44 S & T BANK 9,1 0.7

(Continue)

165

DETERMINANTS OF THE PROFITABILITY OF SAVING S BANKS IN THE US AND THE MODERATING EFFECT OF THE PANDEMIC CRISIS

REVISTA DE ECONOMÍA MUNDIAL 68, 2024, 141-166

45 SPENCER SAVINGS BANK, SLA 3,8 0.3

46 STERLING BANK & TRUST, FSB 2,4 0.2

47 TFS FINANCIAL CORP 15,8 1.2

48 THIRD FEDERAL SAVINGS & LOAN ASSOCIATION OF CLEVELAND 16,1 1.3

49 TIAA, FSB 39,4 3.1

50 TRUSTCO BANK 6,0 0.5

51 UNION SAVINGS BANKS 3,3 0.3

52 UNION SAVINGS BANKS SUBSIDIARES 3,0 0.2

53 UNITED FIDELITY BANK, FSB 5,7 0.4

54 USAA FEDERAL SAVINGS BANK 110,9 8.7

55 WASHINGTON FEDERAL BANK 21,6 1.7

56 WEBSTER BANK NA 71,2 5.6

57 WILMINGTON SAVINGS FUND SOCIETY, FSB 19,8 1.5

TOTAL ASSETS YEAR 2022 - SAMPLE CONSIDERED 920.8 71.9

SAVINGS BANKS NOT CONSIDERED (522) 360.7 28.1

TOTAL BANKING SYSTEM IN THE U.S.A. - SAVINGS BANKS 1.281.5 100.0

APPENDIX II | COMPOSITION OF THE FOUR LINEAR REGRESSION MODELS ESTIMATED

Model 1

RoAit = β0 + β1DEFit + β2LIQit + β3SOLit + β4PROit + DUMMYβ5CRIit + β6INFit + β7CBRit + β8TGOit + β9DIMit + εit

Model 2

RoAit = β0 + β1DEFit + β2LIQit + β3SOLit + β4PROit + DUMMYβ5CRIit + β6INFit + β7CBRit + β8TGOit + β9DIMit +

β10DEFit * CRIit + β11LIQit * CRIit + β12SOLit * CRIit + β13PROit * CRIit + εit

Model 3

RoEit = β0 + β1DEFit + β2LIQit + β3SOLit + β4PROit + DUMMYβ5CRIit + β6INFit + β7CBRit + β8TGOit + β9DIMit + εit

Model 4

RoEit = β0 + β1DEFit + β2LIQit + β3SOLit + β4PROit + DUMMYβ5CRIit + β6INFit + β7CBRit + β8TGOit + β9DIMit +

β10DEFit * CRIit + β11LIQit * CRIit + β12SOLit * CRIit + β13PROit * CRIit + εit

Where,

RoAit represents the return on assets ratio of institution i at time t;

RoEit represents the return on equity ratio of institution i at time t;

DEFit represents the default rate on loans to customers of institution i at time t;

LIQit represents the ratio of deposits to loans of institution i at time t;

SOLit represents the solvency ratio of institution i at time t;

PROit represents the production of complementary activity of institution i at time t;

CRIit represents dummy of the pandemic crisis of institution i at time t;