CC BY
16
BANKING ON MULTINATIONALS: THE DETERMINANTS OF CROSS-BORDER CREDITS TO CENTRAL AND EASTERN EUROPE, 1990-2015
Roksolana ZAPOTICHNA1
Ivan Franko National University of Lviv, Ukraine
Abstract. Taking into account the rising dependence of Central and Eastern Europe on multinational banks' credits, the main purpose of the article is to identify and examine the determinants of cross-border credits through a methodology based on "push" and "pull" factors. The author presents the results of a regression analysis on the determinants of cross-border credits provided by multinational banks from EU-15 to Central and Eastern Europe over the period of 1990-2015 by using the statistical data compiled and published by the Bank for International Settlements. The obtained results suggest that global as well as home and host country level determinants influence cross-border credits but to a different extent. According to our results, higher stability and predictability of global economic environment contribute to higher cross-border credit growth. The results of the host country determinants analysis indicate that more effective and profitable economies receive more credits from multinational banks. We also find that multinational banks provide more credits to countries with small markets, low inflation rate, high external debt burden, high capital account deficit, fixed exchange rate regime, and developed institutional environment. Regarding home level determinants, we find the existence of a negative correlation between home country economic cycle and the amount of cross-border credits received by the host country, which can be explained by low economic growth in continental Europe over the period under consideration that stimulated European banks to expand lending on foreign markets with higher profit opportunities. Thus, cross-border credits appear to have been countercyclical to growth in home countries and procyclical to growth in host countries. Finally, it is found that host country level determinants play the most important role in explaining changes in cross-border credits on host countries in Central and Eastern Europe during 1990-2015 period, which means that host countries have a control over their own destinies and the amount of received credits depends on their economic and political performance. Value/originality. The results of the research make it possible to provide a better understanding of the determinants of cross-border credits and practical importance of multinational banks' lending as an important source of external finance for the catching-up process and a major component in the ongoing process of financial deepening in Central and Eastern Europe, and clarify whether these determinants differ in periods of financial stability and crisis.
Key words: Multinational banks, cross-border credits, Central and Eastern Europe, push factors, pull factors, regression analysis.
JEL Classification: F23, F37, G21, N24
1. Introduction
One of the key peculiarities of global financial market activity over the 1990s was the dramatic growth in multinational banks' credits to emerging markets, including the countries of Central and Eastern Europe. Essentially, multinational banks provide credits to host countries via two different ways: directly, from their parent banks abroad (so-called cross-border credits) or indirectly, through local subsidiaries or branches in host countries (Kamil & Rai, 2010). Given that most of the 1990s increase in multinational banks' credits to Central and Eastern Europe was related to the establishment of
branches and subsidiaries, most existing studies were focused on the indirect credits: their determinants and impact on the host countries. The literature on direct cross-border credits has received comparatively less attention so far, mainly because of data limitations. Although cross-border credit is an important source of external finance for the catching-up process and a major component in the ongoing process of financial deepening, understanding the driving forces of these credits in Central and Eastern Europe is of particular importance. The determinants of cross-border credits should be accurately considered when examining the transmission mechanism of financial distress from
Corresponding author:
1 Department of International Economic Analysis and Finance, Ivan Franko National University of Lviv. E-mail: roksolana.zapotichna@gmail.com
advanced to emerging markets, their consequences for the financial stability of the host countries. Cross-border credits can be characterized by periods of rapid growth followed by a sharp decline. This so-called "boom and bust cycle" may have devastating consequences for countries highly dependent on cross-border credits. This is especially the case for the countries of Central and Eastern Europe, which are significantly exposed to the banks from EU-15.
Researches (Jeanneau & Micu, 2002, Khattak, 2011) classify the determinants of cross-border credits by employing traditional "push vs pull" framework. The changes in the amount of credit provided by multinational banks in response to changes in economic conditions in home country and global environment are termed as "push factors" (in other words, external factors outside the control of a specific borrowing country), and the variation in the amount of credit in response to host country conditions of economic and non-economic nature referred to as "pull factors".
Depending on the study, push and pull factors are sometimes branded as "supply-side" and "demand-side" (Amiti, McGuire & Weinstein, 2016).
The objective of the article is to identify the main determinants of cross-border credits in Central and Eastern Europe relying on the regression analysis. The implemented regression model is based on the existing researches on multinational banking and on the extensive literature on capital flows, which is too numerous to review here. Central and Eastern Europe have been chosen for the objective of our analysis for at least three reasons. First, the choice was made based on the existence and availability of comparative data base. Second, this group of countries was deliberately chosen because of its heterogeneity. The region of Central and Eastern Europe includes countries with fixed and floating exchange rate regimes, with various levels of financial liberalization and economic development, which differ from each other in the amount and importance of cross-border credits.
Our choice of home countries was driven by their relative importance as lenders to Central and Eastern Europe. We consider separate credits from multinational banks based in EU-15, which account for nearly 77% of all credits provided by BIS-reported banks to the region, and when taken together are the most important lenders to Central and Eastern Europe.
The period of 1990-2015 was chosen because during this time frame foreign financing rose sharply in the region. Moreover, during this time countries of Central and Eastern Europe have liberalized their financial systems, which led to the reorientation of multinational banks from almost purely cross-border lending to a mix that also included indirect lending.
Our contribution to the existing literature on cross-border credits is as follows: we examine the determinants of cross-border credits on longer time period than in
some of the previous studies; we investigate whether the explanatory power of different groups of determinants has changed over the time period under consideration by splitting it into two subperiods. This, in turn, will help to clarify whether these determinants differ in periods of financial stability and crisis.
2. Data and methodology
Our data on cross-border credits come from Consolidated banking statistics compiled and published by the Bank for International Settlements. To the best of our knowledge, data published by the Bank for International Settlements is the most comprehensive data available that is well suited to an analysis of the determinants of cross-border lending since it provides information about the nationality and location of the lending banks and borrowers.
Consolidated banking statistics include cross-border credits, local credits of the foreign affiliates in foreign currencies, and local credits of these affiliates in the local currency. Cross-border credits and local credits in foreign currencies are reported by the Bank for International Settlements as a single inseparable series only. Local credits in local currency are reported as a separate series. Given that Consolidated banking statistics is biannual until 2000 and quarterly thereafter, data availability makes us focus on annual, end-of-year statistics.
The data for explanatory variables come from different sources: International Financial Statistics (IMF), Political Risk Index Data (PRS Group), World Development Indicators (World Bank), the joint BIS-IMF-OECD-World Bank statistics on external debt.
Panel data techniques were used instead of separate time series and cross-border sections in order to collect more information and to overcome potential estimation biases resulting from possible correlations between regressors and residuals.
Following the approach used by Weller (2001), we chose the ratio of cross-border credits to GDP as the dependent variable. While using the "push vs pull" framework, we divided the independent variables into three groups: global level, home country level, host country level. The host country level determinants were further divided into determinants of economic and non-economic nature.
The expected sign of the coefficients of variables, their indicators and abbreviation are presented in Table 1.
Hence, the basic regression equation has the following specification:
MNBloansit V V V
-GD^-= a + Pi Hostit + ^ P2 Home;, + ^ P3 Global, + ;, (1)
where
MNBloansit/GDPit - is the ratio of cross-border credits to GDP;
Hostit - is a matrix of host country macroeconomic and non-economic variables;
Vol. 3, No. 1, 2017 Table 1
Description of the explanatory variables
Variable Indicator and abbreviation Expected sign
Global level
Uncertainty in
global financial S&P500 index (S&P500) +
markets
Home country level
Economic growth GDP growth level (GDP Home) +/-
Real interest rate (Interest Home) -
Host country level
Non-economic nature
Voice and accountability (VA);
Political stability and absence of violence (PV);
Political stability Government effectiveness (GE); Regulatory quality (RQ); Rule of law (RL); Control of corruption (CC). +
Economic nature
Economic
growth/demand GDP growth (GDP) +
for MNBs' credits
Economic GDP per Capita growth
efficiency (GDPperCap)
Future economic growth Foreign direct investment to GDP ratio (FDI) +/-
Market size Population (Population) +
Interest rate (Interest) +
Inflation (Inflation)
Exchange rate (ER) -
Exchange rate regime Reinhart-Rogoff index (EXregime) -
External debt External debt to GDP ratio (Debt) +/-
Financial
openness of the Chinn-Ito index (Open) +
banking sector
Current account balance Current account to GDP ratio (CAB) +/-
Source: created by author
Homejt - is a matrix of home country macroeconomic variables;
Globalt - is a matrix of the global level variables; j - identifies home countries;
i - indicates each individual Central and Eastern European host country; t - refers to the time period considered; sijt - are stochastic disturbances. All variables are expressed in percentage points, except S&P500 index and population, which are expressed in logarithms. Reinhart-Rogoff index ranges from 1 to 3: higher, more flexible exchange rate regime (Reinhart & Rogoff, 2004). Chinn-Ito index takes on higher values the more open the country is to cross-border capital transactions (Chinn & Ito, 2007). Political stability
indicators range from 0 to 1, where a higher value indicates more stable political environment.
It is also essential to examine if there is evidence that the determinants of cross-border credits have changed over the period under consideration. To this purpose, we estimated our basic equation over two subperiods: 1990-2002, when a number of cross-border credits were in general low and the credit activity of foreign banks was in some cases forbidden; and 2003-2015, when Central and Eastern European economies experienced credit boom. Specifically, we explore whether credits' sensitivity to global, host and home conditions has changed over time. As mentioned by Aysun and Hepp (2016), if host country factors are more important, this could suggest that a host country with high and sustainable growth would receive more credits from multinational banks and that these credits would decrease in an economy, which is performing poorly. As a consequence, countries would be in charge of their own destinies. And vice versa, if global and home country factors are more important, the state of an economy may be less related to the credits it receives and economies may be more responsive to external developments
In order to carry out this study, the research questions were broken down into several empirically testable hypotheses. Taking into account the theoretical background the following hypotheses were formulated:
1. Cross-border credits provided to a specific host country are expected to have a negative correlation with home country economic performance.
2. Cross-border credits are expected to have a positive relation to global developments as well as economic and non-economic conditions in a host country.
3. Cross-border credits' sensitivity to different groups of determinants has changed over time.
3. Empirical results
The results of our baseline regression equation are presented in Table 2.
We estimated the impact ofthree groups of explanatory variables on cross-border credits both jointly and separately, because estimated determinants may not be completely independent of each other. After that, we compared these models in terms of their significance by looking at the coefficients of determination R2 and the F-statistics. The estimated variables are robust with respect to different model specifications, so we presented the results of one large model in order to avoid double notation.
Altogether, most estimated variables have the expected signs and are statistically significant. The regression analysis shows that global as well as home and host country level determinants are significant driving forces of cross-border credits. Our estimations are in line with the results obtained in previous studies. For example,
Baltic Journal of Economic Studies , -J--Vol. 3, No. 1, 2017
Table 2
The results of the regression analysis on the determinants of cross-border lending
Indicators Time period
1990-2015 1990-2002 2003-2015
Log(S&P500) 18,816 (9,876)*** 18,861 (6,700)*** 17,856 (5,956)***
GDP Home -3,387 (-3,473)*** -2,468 (-2,813)** -1,998 (-2,351)**
Interest Home -6,654 (-8,298)*** -5,296 (-4,657)*** -4,961 (-4,184)***
VA 94,208 (2,678)** 36,469 (0,575) 21,145 (0,286)
PV -38,511 (-2,145)* -38,002 (-1,082) -24,721 (-0,525)
GE 57,386 (3,597)*** 34,218 (1,004) 28,501 (0,708)
RQ 20,253 (1,717) 17,305 (1,321) 14,091 (0,823)
RL 55,085 (3,358)*** 94,701 (4,257)*** 93,314 (3,691)***
CC 2,263 (1,316) -35,556 (-1,386) -41,378 (-1,486)
GDP 0,155 (2,711)** 1,866 (2,592)* 1,798 (2,531)*
GDPperCap -1,316 (-3,308)** -2,091 (-0,655) -1,952 (-0,566)
FDI 0,417 (4,407)*** 0,582 (3,969)*** 0,642 (3,412)***
Log(Population) -15,599 (-6,636)*** -15,666 (-3,491)*** -16,962 (-2,943)**
Interest 0,224 (1,521) 0,325 (1,559) 0,321 (1,389)
Inflation -0,445 (-3,081)** -0,757 (-3,098)** -0,6886 (-2,301)*
ER 0,063 (1,999)* 0,025 (0,595) 0,039 (0,784)
EXregime -11,375 (-4,689)*** -18,486 (-4,283)*** -15,574 (-2,654)**
Debt 0,232 (3,976)*** 0,346 (5,029)*** 0,359 (2,732)**
Open 2,488 (1,787) 2,648 (1,134) 2,138 (0,781)
CAB -1,103 (-3,634)*** -0,914 (-2,643)** -0,814 (-2,013)*
Number of observations 250 130 120
R2 0,86 0,951 0,952
Adjusted R2 0,83 0,923 0,917
F-statistics 28,18 33,97 27,85
Source: The authors' calculations
Notes: Standard errors are robust. T-statistics appear in parentheses and ***, **, * correspond to the 1%, 5%, and 10% level of significance, respectively
like Takats (2010) and Ghosh (2011), we find that higher stability and predictability of global economic environment, measured by S&P 500 index, promotes cross-border credit growth. We also experimented with alternative global level indicators, such as world real GDP, real US interest rates, and trade volumes growth, but the results to a great extent remained unchanged.
Regarding home level determinants, our estimations confirm the results of the researches by Molyneux and Seth (1998), Haas and Lelyveld (2008) who find the existence of a negative correlation between home country economic cycle and a number of cross-border credits received by the host country. The countercyclical behaviour of cross-border credits provided by the banks from EU-15 could largely be explained by the fact that economic growth in continental Europe was generally low over the period under consideration, lending opportunities and returns in domestic markets were probably limited, driving European banks to expand lending on foreign markets with higher profit opportunities. Indeed, banks based in EU-15 were the most active lenders to Central and Eastern Europe in the 1990s. All in all, the findings from splitting the sample into the periods of pre- and post- 2002 suggest that in recent years the behaviour of banks from EU-15 tends to be less countercyclical to home country economic growth.
The results of host country determinants analysis are in line with the studies, which find out that more effective and profitable economies with higher growth perspective receive more credits (Focarelli and Pozzolo, 2004, for instance). Among the subset of host country determinants, we find that the coefficient on host country GDP growth is positive and significant, showing that multinational banks respond to host country growth, increasing and decreasing credits over the cycle. Multinational banks also direct their credits to countries with optimistic growth perspective, which is measured by foreign direct investment to GDP ratio. Besides, the positive sign of the coefficient indicates that cross-border credits and foreign direct investments are rather complementary than substitutable sources of external financing. The results of the regression analysis indicate that multinational banks provide more cross-border credits to countries with small markets, low inflation rate (which is consistent with the findings obtained by Focarelli and Pozzolo, 2004, Derviz and Podpiera, 2007), fixed exchange rate regime (which is in line with results obtained by Jeanneau and Micu, 2002). Additionally, countries with capital account deficit receive more cross-border credits. Thus, higher current account balance in the past means that less borrowing is required to finance the current account. We also find that multinational banks provide more credits to countries with higher external debt to GDP ratio: the higher the debt of a country relative to the size of the economy, the greater the need to borrow from
abroad. With regard to determinants of non-economic nature, the only significant variable for all periods under consideration is regulatory quality. The positive coefficient on this determinant indicates that countries with better regulatory quality receive more credits from multinational banks.
Using the methodology developed be Goldberg and Saundres (1981), we also experimented with the combinations of host and home level variables, as cross-border credits may not so much respond to home or host country conditions, but rather to the difference between them. Accordingly, we included two indicators: host minus home country GDP and host minus home country interest rate. However, the coefficients on those determinants appeared to be insignificant, and as the result were not included in the final regression equation.
As a next step, we analysed the explanatory power of three groups of explanatory variables. This analysis goes beyond the identification of statistically significant determinants of cross-border bank credits and provides additional information on the economic significance of estimated parameters. The contribution of each variable is calculated by using the methodology applied by Martinez Peria, Powell, and Vladkova-Hollar (2005), Zhu, and Decady (2007) who calculate the percent variance explained as (R2 _full-R2 _constrained)/R2 _full *100. In other words, for each group of variables, they compute the increase in the R2, as a proportion of the total variance of the percentage change in credits explained by all variables. The results of our calculations are presented in Table 3.
Table 3
Explanatory power of global, home and host country level determinants of cross-border credits
Group of determinants Time jeriod
1990-2002 2003-2015
Global level 18,49% 17.29%
Home country level 14,07% 14,28%
Host country level 67,44% 68,43%
Source: authors' calculations
The figures in Table 3 show that host country level determinants explain almost twice as much change in cross-border credits as global level and home country level taken together. Additionally, the explanatory power of all three groups of determinants has remained largely unchanged in both periods under consideration. These findings confirm that host countries are responsible for the amount of credit they may receive from multinational banks.
4. Conclusions
Given that tendency toward multinational banks' credit expansion in emerging markets is likely to continue, it is important to study the determinants and
the consequences ofthis expansion for the host countries. In this article, we considered the Bank for International Settlements statistics and regression analysis techniques to investigate the aforementioned issues on the example of Central and Eastern Europe, a region that witnessed a substantive increase in multinational banks financing during the transition period of the 1990s.
Our results confirm two out of three hypotheses, which have been formulated in the previous section of the article: cross-border credits appear to be countercyclical to growth in EU-15, which have been chosen as home countries for the objectives of our research, and procyclical to growth in Central and Eastern Europe, which have been chosen as host countries. Additionally, the obtained results suggest that all three groups of determinants influence cross-border credits but to different extent. It is found that host country level determinants play the most important role in explaining changes in cross-border credits on
References:
host countries in Central and Eastern Europe during 1990-2015 period, which means that host countries have a control over their own destinies and a number of received credits depends on their economic and political performance. This conclusion has important policy implications which may be the subject for further researches.
It is also important to mention that treating multinational banks as a homogenous group might be misleading; therefore, multinational banks' heterogeneity should be taken into account while examining their cross-border lending in future studies.
Moreover, the data used in this article have some restriction. For example, the Bank for International Settlements statistics does not provide separate data on "pure" cross-border credits and local credits in foreign currency, which may bias the results of our research. Although, the analysed time period is longer than in some of the previous studies, but still is relatively short.
Amiti, M., McGuire, P., Weinstein, D. (2016). "Supply- and Demand-Side Factors in Global Bank Credit". [Electronic resource]. - Retrieved from: http://ijournales.org/archive/volume2016/issue5/10.pdf Aysun, U., Hepp, R. (2016). The determinants of global bank lending: Evidence from bilateral cross-country data, Journal of Banking & Finance, Volume 66, pp. 35-52.
Bank for International Settlements. Official site. [Electronic resource]. - Retrieved from: https://www.bis.org/ Buch, C., Lipponer, A. (2004). FDI versus cross-border financial services: The globalisation of German banks, Deutsche Bundesbank Discussion Paper, Series 1: Studies of the Economic Research Centre, No 05, 52 p. Chinn, M., Ito, H. (2007). A New Measure of Financial Openness. [Electronic resource]. - Retrieved from: http://www.web.pdx.edu/~ito/kaopen_Chinn-Ito_hi0523.pdf
De Haas, R., Van Lelyveld, I. (2008). Internal capital markets and lending by multinational bank subsidiaries, EBRD Working Paper No. 105, p. 35.
Derviz A., Podpiera J. (2007). Cross-Border Lending Contagion In Multinational Banks, European Central Bank Working Paper Series, No. 807, 40 p.
Ghosh, N. R., Sugawara, N., Zalduendo, J. (2011). Banking Flows and Financial Crisis Financial. Interconnectedness and Basel III Effects, Policy Research Working Paper, 5769, 28 p.
Goldberg, L.G., Saunders, A. (1981). The determinants of foreign banking activity in the United States, Journal of Banking & Finance 5, pp. 17-32.
International financial statistics. Electronic resource. Retrieved from: https://www.imf.org/en/Data
Jeanneau, C., Micu, M. (2002). "Determinants of International Bank Lending to Emerging Market Countries", BIS Working Papers, 112, 43 p.
Joint BIS-IMF-OECD-World Bank statistics on external debt. [Electronic resource]. - Retrieved from: http:// www.bis.org/publ/r_debt.htm
Kamil, H., Rai, K. (2010). "The Global Credit Crunch and Foreign Banks' Lending to Emerging Markets: Why Did Latin America Fare Better?", IMF Working Paper, 102, 33 p.
Khattak, M. (2011). "Impact of Foreign Bank Entry on the Credit Stability of Host Countries. A Study on South Asia". [Electronic resource]. - Retrieved from: http://aut.researchgateway.ac.nz/bitstream/handle/10292/3339/ KhattakM.pdf?sequence=3
Martinez Peria, M., Powell, A, Vladkova-Hollar, I. (2005). Banking on Foreigners: The Behavior of International Bank Claims on Latin America, 1985-2000, IMF Staff Papers Vol. 52, Number 3, 32 p.
Molyneux, P., Seth, R. (1998). Foreign Banks, Profits and Commercial Credit Extension in the United States, Applied Financial Economics, 8(5), pp. 533-539.
Political Risk Index Data. Electronic resource. Retrieved from:https://www.prsgroup.com/category/risk-index Reinhart, C., Rogoff, K. (2004). The Modern History of Exchange Rate Arrangements: a Reinterpretation, The Quarterly Journal of Economics, Vol. CXIX, Issue 1, 48 p.
Takats, E. (2010) Was It Credit Supply? Cross-Border Bank Lending to Emerging Market Economies During the Financial Crisis, BIS Quarterly Review, pp. 49-56.
Thomas, D. R., Zhu, P., Decady Y. (2007). Point estimates and confidence intervals for variable importance in multiple linear regression, Journal of Educational and Behavioral Statistics, 32, pp. 61-91.
Weller, C. (2001). The Supply of Credit by Multinational Banks in Developing and Transition Economies: Determinants and Effects, DESA Discussion Paper, No. 16, 30 p.
World Data Bank, World Development Indicators. Electronic resource. Retrieved from: http://databank.worldbank.org/data
Роксолана ЗАПОТИЧНА
ДЕЯТЕЛЬНОСТЬ ТРАНСНАЦИОНАЛЬНЫХ БАНКОВ: ДЕТЕРМИНАНТЫ ТРАНСГРАНИЧНОГО КРЕДИТОВАНИЯ СТРАН ЦЕНТРАЛЬНО-ВОСТОЧНОЙ ЕВРОПЫ
Аннотация. Принимая во внимание растущую зависимость стран Центральной и Восточной Европы от кредитов транснациональных банков, основная цель статьи - выявить и изучить детерминанты трансграничных кредитов с помощью методологии, основанной на "пуш" и "пулл" факторах. Приведены результаты регрессионного анализа детерминантов трансграничных кредитов, предоставленных транснациональными банками из 15 стран ЕС странам Центральной и Восточной Европы за период 19902015 гг., с использованием статистических данных, собранных и опубликованных Банком международных расчетов. Полученные результаты позволяют предположить, что глобальные детерминанты, детерминанты на уровне страны происхождения, так и на уровне принимающей страны влияют на международные кредиты, но в разной степени. Согласно нашим результатам, более высокая стабильность и предсказуемость глобальной экономической среды способствуют более высокому росту трансграничного кредитования. Результаты анализа детерминантов на уровне принимающей страны свидетельствуют о том, что более эффективные и прибыльные экономики получают больше кредитов от транснациональных банков. Мы также сделали вывод о том, что транснациональные банки предоставляют больше кредитов странам с небольшими рынками, низкой инфляцией, высоким бременем внешней задолженности, высоким дефицитом счета операций с капиталом, режимом фиксированного обменного курса и развитой институциональной средой. Что касается детерминантов на уровне страны происхождения, мы обнаруживаем наличие отрицательной корреляции между экономическим циклом страны происхождения и количеством трансграничных кредитов, полученных принимающей страной, что можно объяснить низким уровнем экономического роста в континентальной Европе за рассматриваемый период, что стимулировало европейские банки расширяют кредитование на внешних рынках с более высокими возможностями получения прибыли. Таким образом, трансграничные кредиты, как представляется, являются антициклическими к росту в странах происхождения и проциклическими к росту в принимающих странах. Наконец, было установлено, что детерминанты на уровне принимающей страны играют самую важную роль в объяснении изменений в трансграничных кредитах в странах Центральной и Восточной Европы в период 1990-2015 годов, а это означает, что страны пребывания контролируют свои судьбы, и сумма полученных кредитов зависит от их экономических и политических показателей. Значение/оригинальность. Результаты исследования позволяют лучше понять детерминанты трансграничных кредитов и практическую важность кредитования транснациональных банков в качестве важного источника внешнего финансирования и основного компонента текущего процесса финансового углубления в Центральной и Восточной Европе и уточнить, отличаются ли эти детерминанты в периоды финансовой стабильности и кризиса.
