Abstract
In this article we analyze the Hungarian shadow banking system. We point out that the Hungarian shadow banking system is not only much less developed than that of the EU's developed countries, but also structurally different. A further specific feature of the Hungarian financial system is what we call the secondary shadow banking system, through which foreign shadow banking funds do not finance the domestic banking system directly, but through foreign interbank funds and related cross currency basis swaps. The aim of our analysis is to explore the reasons for these specificities, to analyze the risks of the Hungarian shadow banking, and secondary shadow banking systems, and to show that the interconnectedness between banking and shadow banking may not only occur through direct exposure, but also indirectly through the presence of secondary shadow banking.
1 Introduction
According to the commonly used definition, shadow banking (SB) is a segment of credit-based financial intermediation that involves financial intermediaries and financial market segments outside the traditional commercial banking system (FSB 2011). The emergence of the SB system, its inherent risks, and the regulatory arbitrage that took place within it, contributed greatly to the emergence and internationalization of the financial crisis that started in the US in 2007 (Pozsar et al. 2010; Adrian – Ashcraft 2012; Lysandrou – Nesvetaliova 2015; Ban – Gabor 2016; Thiemman 2018). The process also transformed the business model of many banks that previously operated in a traditional way, i.e., by collecting deposits and granting loans, and holding the loans in the portfolio until maturity (originate-and-hold). The so-called originate-and-distribute banking model emerged, where banks originated loans for securitization and market sale.
The SB system spread to Europe partly by buying assets created by securitizing loans in the US and partly by adopting the business model based on securitization. In addition, there was a trend in Europe that banks increasingly funded their long-term lending with short-term non-deposit liabilities raised from the market. Thus, increased reliance on capital market investment (on the assets side) and market-based funding (on the liabilities side) became widespread.
These trends have also led to the convergence of traditional banking, capital markets and SB in Europe. This process is referred to as market-based banking (Hardie et al. 2013).
Securitization has not taken root in East-Central Europe, including Hungary. Only a limited number and volume of transactions have taken place. Market participants in the region made only very limited purchases of financial assets that proved to be “toxic” during the crisis. In 2007, therefore, it seemed that the countries of the region might be spared the crisis, which initially spread through the SB channels. After 15 September 2008, following the collapse of Lehman Brothers, this hypothesis was quickly disproved (Dooly – Hutchinson 2009; Király – Mérő 2010). Although SB instruments and institutions were not the main channel of contagion in the region in the 2008 global financial crisis, the institutions and activities that are part of the SB system emerged nonetheless (Bethlendi – Mérő 2020a, 2022). The main contagion channel of the crisis was that in several countries of the Central and Eastern European (CEE) region, including Hungary, the pre-crisis credit boom took the form of foreign currency lending. Foreign currency loans, including long-term mortgages, were financed by short-term loans from foreign banks, especially the parent banks of domestic banks, rather than by domestic deposits or domestic capital market funds (Banai et al. 2010). However, these funds came largely from the home capital markets of the banks providing the funds, in accordance with the rise of market-based banking on their home markets. In other words, deposits were increasingly replaced by SB funds raised from the markets and intermediated through foreign interbank lending. The market-based nature of funding was further reinforced by the fact that these funds were often associated with cross-currency basis swaps (Bethlendi 2015).
Thus, in Hungary, exposure to the SB system was not only direct, but also indirect, through foreign banks, as a source of market-based funding. We term this phenomenon the “secondary shadow banking system”. The idea behind this approach is based on Harutyunyan et al. (2015), who measure the shadow banking system by the non-core liabilities of financial institutions. Accordingly, in this paper we use two different approaches to analyze the Hungarian domestic SB system. First, we analyze it from the perspective of direct SB functions and institutions and, second, from the perspective of the emergence of a secondary SB system. Our aim is to present and analyze the specific structure that characterized the Hungarian financial system in the period before and after the 2008 crisis: namely, the pre-crisis emergence of the secondary shadow banking system alongside the relatively underdeveloped shadow banking system, which made the banks' funding structure particularly vulnerable. This twofold approach to shadow banking is our main contribution to the existing literature, since, using the Hungarian example the article sheds light on a new, hitherto unexamined feature of the SB system.
The structure of the article is as follows. The second section contains a literature review on shadow banking in Europe with special regard to the CEE region. The third section provides an overview of our data and methodology. In the fourth section we describe the structure, size, the main features and risks of the Hungarian shadow banking system. The evolution and characteristics of the secondary shadow banking system are analyzed in the fifth section. The last section contains the conclusions.
2 Literature review
The emergence and the general characteristics of SB, as well as its risks are well researched topics (Pozsar et al. 2010; Adrian – Ascraft 2012; Acharya et al. 2013; Claessens – Ratnovski 2014; Thiemann 2018; Thiemann – Tröger 2021). Prior to the Global Financial Crisis (GFC), the US SB sector was significantly larger than that of the euro area. After the GFC, the US SB sector started to shrink before regaining its strength in the mid-2010s. In Europe, the SB sector continued to increase even during the early 2010s (Pires 2019). As a result, by 2016 the euro area SB system was somewhat larger than that of the US (FSB 2020, Graph 1–4, p. 12). However, due to the bank-based nature of the European financial system, compared to traditional bank assets SB assets still played a less significant role in European financial intermediation than in that of the US. Accordingly, despite the significant growth of the European SB system, the financial system in Europe remained less exposed to SB developments (Bakk-Simon et al. 2012; Malatesta et al. 2016). As factors for the growth of EU SB – using dynamic panel regression – Hodula et al. (2020) identify real GDP growth, the demand of institutional investors, tightening capital requirements of banks and the financial development of EU member states. The interconnectedness of SB and other parts of the financial system for the euro area is analyzed by Giron and Matas (2017). Using the ECB's who-to-whom data, they analyze the flow of funds between different categories of financial institutions. The interconnectedness of EU banks and SB has been analyzed by Abad et al. (2017). They highlight the high interconnectedness of EU banks and the global SB, especially that of the US.
Hodula (2022) compares the SB system of old and new EU member states, where the category of new member states includes, with the exception of Croatia, the CEE EU member states, as well as Cyprus and Malta. His main finding is that in new member states SB is relatively simple and it is more of a substitute for banking. In old member states, SB is more complex and it complements, rather than substitutes, traditional banking.
The CEE countries' SB itself has been subject to relatively few analyses. Ghosch et al. (2012), focusing on the SB of emerging markets, include three CEE countries, namely Bulgaria, Romania and Croatia, for the period 2003–2011. They stress that despite the low level of SB in these countries, the substantial increase of SB activities in recent years, and its interconnectedness with the traditional financial system, could also cause a systemic risk in these countries. Measuring the CEE SB, Bethlendi and Mérő (2022) point out that Hungary's depth of financial intermediation, both in terms of banking and shadow banking activities, is in the middle of the region, significantly below the level of developed countries. In line with the underdeveloped capital markets of the region, the structure of SB is also different: investment funds play a smaller role, and financial corporations engaged in lending (FCLs), i.e., non-bank factoring- leasing- and lending corporations play a much bigger role.
Apostoaie and Bilan (2020) use panel regression to analyze the macro determinants of the dynamics within the shadow banking system of the 11 CEE countries for the period between 2004 and 2017. They found that from a low basis in 2004, the SB of the CEE countries had grown more considerably than that of the Eurozone countries between 2004 and 2007. They identified economic growth, the state of global liquidity, the search for yield of market investors, demand from institutional investors, and the complementarity of SB system to the rest of the financial system as the most significant macro-drivers of the process. The latter means that traditional banking is not able to provide alternative funding for lending, while SB is.
Kjosevski et al. (2021) also analyze the drivers of SB growth in CEE for 1999–2019. They also identified economic growth as the main driver of SB. Their analysis divided the 11 CEE countries into three, more homogenous groups, namely the Baltic states, the Visegrad countries, and the Balkan countries. According to their results, the SB of the sub-groups differed most in how the global financial crisis affected them. They demonstrate that the Hungarian SB to GDP ratio is several times higher than that of other countries of the region. However, they use the Eurostat database for the analysis, which may be misleading for Hungary because the inclusion of captive financial institutions highly inflates the Hungarian figures (see next Section on data).
3 Data and methodology
From an institutional perspective, SB includes money market funds, non-money market funds, and other financial intermediaries. The latter is an aggregated category, of which the largest part is made up of FCLs, namely lending-, leasing-, and factoring companies.
For the structural analysis of SB, we use the annual data published by Bethlendi and Mérő (2022) on SB of the CEE countries, including Hungary, for the period 2004–2019. The database contains data on the size of each subsector of the SB system. As a starting point, these data are built on the Eurostat financial accounts and monetary statistics. However, systematic data cleaning and data correction were carried out. On the one hand, several country-specific data were added to their database. On the other, significant data exclusion was made as well, that of captive financial firms. The latter is in line with the Financial Stability Board (FSB) methodology, but is not matched with Eurostat data, so the data cleaning can only be done individually. In most countries the size of the captives is negligible, but in some countries, such as Hungary, it is extremely large. In Hungary the captives are not engaged in financial markets (Koroknai – Lénárt-Odorán 2011). Another important data correction they have made, also in line with the FSB methodology, is that the volume of investment fund assets has been reduced by the volume of equity investments, which cannot be part of the credit intermediation.
The secondary SB analyses is based on the distinction between core- and non-core liabilities of traditional commercial banks. This includes liabilities raised from money and capital markets, whatever form they take (deposits, borrowing, debt securities issuance). According to Harutyunyan et al. (2015) interbank and central bank borrowing are excluded from the non-core liabilities on the ground that they are used for the proper day-to-day functioning of markets (liquidity management) and not as a source of credit financing. In the Hungarian (and generally CEE) cases, this only applies to domestic interbank loans. In most countries of the region, the role of foreign bank finance is not to manage temporary liquidity needs and surpluses of banks, but to increase the lending potential. Consequently, for the purpose of analyzing the Hungarian case, foreign interbank lending to Hungarian banks is classified under the category of non-core liabilities. This approach is in line with Shin-Shin (2011), which, in case of open economies, argues in favor of defining foreign interbank funds as non-core liabilities, as they behave similarly to market funds.
To analyze the secondary SB system, we need a detailed breakdown of banks' liabilities to distinguish between core and non-core liabilities. In the analysis of the funding structure of the Hungarian banking system, the following items are classified as non-core liabilities: 1) liabilities raised from non-bank financial- and investment corporations and mutual funds (excluding money market funds); 2) money market funds; 3) liabilities raised from insurance companies and pension funds; 4) debt securities issued by banks; and 5) foreign interbank liabilities.
Categories 1 to 4 are collectively referred to as market funds, which in this approach is the direct SB funding of banks.1 In accordance with our concept of SB, the secondary SB funds raised by banks are foreign interbank funds (Category 5). The combined evolution of these two categories is an indicator of the market-based nature of the Hungarian banking sector.
The Hungarian central bank (Magyar Nemzeti Bank, MNB) provides data on banks' liability structure up to 2016 with the level of detail required for the analysis. Accordingly, the time horizon for this part of the analysis is 2004–2016. Starting from 2017, in line with the gradual introduction of IFRS accounting standards, the data on banks' liabilities structure for banks reporting under IFRS has been reduced significantly, from 300+ items to 34 items, which no longer allows for separation of core- and non-core liabilities. Most of the large Hungarian banks (OTP, CIB, KH, UniCredit) have already switched to IFRS reporting as of 2017. Thus, even though the previous detailed data set is still available for banks reporting according to Hungarian accounting standards for the transition period, it is not possible to draw conclusions on the liability structure of the banking system after 2017.
Based on these two different datasets and several additional data sources, a structural analysis using descriptive statistical methods of SB and secondary SB of Hungary is carried out.
4 Structure and risks of the Hungarian shadow banking system
In the advanced EU Member States, which are crucial for financial sector ownership in CEE countries, the SB-to-GDP ratio was significantly higher than that of the CEE countries during the whole period under review. Comparing with the Visegrad Group, in 2010 the Hungarian SB-to-GDP was the highest, but by 2019, the Czech ratio was significantly higher, while the Polish one was slightly higher (Fig. 1).
SB-to-GDP (%) in selected advanced EU member states and in the Visegrad Group (%)
Source of data: Bethlendi – Mérő (2022).
Citation: Society and Economy 46, 3; 10.1556/204.2024.00001
4.1 Structural issues
The financing structure of the Hungarian domestic private sector is dominated by domestic bank loans. Corporate bond issue is marginal and domestic equity market capitalization is relatively stagnant. The financing role of SB entities, mainly that of FCLs, is also significant (Table 1, 2nd column). Banks used to issue a significant amount of bonds to refinance their activities, of which mortgage bonds account for about half. We see a general definancialization trend in the 2010s. The role of domestic financial institutions in granting funds to the domestic private sector declined significantly. As bank and SB lending declined, so did the importance of bank bond issues. For comparison, we show in Table 1 the domestic government bond market, which in 2019 had already surpassed the size of the domestic private sector funding.
The domestic financing structure of the non-financial private sector, bank bonds, and government bonds in Hungary (% of GDP)
| (1) Loans from bank | (2) Loans from shadow banks | (3) Corporate bonds | (4) Equity market capitalization | (1)–(4) Non-financial private sector financing* | Bonds issued by banks | Government bonds | |
| 2004 | 42.1 | 7.8 | 0.41 | 22.3 | 72.6 | 9.17 | 61.1 |
| 2008 | 63.7 | 11.8 | 1.59 | 17.6 | 94.7 | 15.62 | 64.9 |
| 2010 | 60.3 | 11.5 | 1.95 | 21.0 | 94.7 | 15.6 | 64.6 |
| 2019 | 33.4 | 5.6 | 1.56 | 20.1 | 60.6 | 5.89 | 73.8 |
*Without intercompany loans and other foreign liabilities.
Source: Hungarian Central Statistical Office, MNB, World Bank, Bethlendi - Mérő (2022).
We are not aware of any market-based securitization of loan portfolios by Hungarian banks. In Hungary, securitization only took place to strengthen the capital position of foreign banks' Hungarian subsidiaries through guarantees by foreign parent banks linked to securitization transactions, but their volume was not significant.2 Neither has repo-based shadow banking3 spread in Hungary. The domestic repo market is almost exclusively based on government securities. During the period under review, banks have almost exclusively entered repo transactions with foreign counterparties, using a part of their government bond portfolio. The available data4 suggest that, even in the most active periods, repo holdings never exceeded 2% of the balance sheet total of banks, and were below 1% for most of the period.
Overall, the shadow banking funding channels that have become a key element of market-based banking have not emerged. The exception is the relatively large volume of FCL lending, but even this has taken place through traditional on-balance sheet financing rather than through securitization.
Although the size of the Hungarian SB system is relatively small compared to developed countries (Bethlendi – Mérő 2022), it was not negligible during the period under review. The composition of the Hungarian SB system by type of institutions is shown in Table 2.
Main institutions of the Hungarian shadow banking system (total shadow banking system* = 100%)
| Money market funds (%) | Non-money market funds (%) | Other financial intermediaries (%) | |
| 2004 | 10 | 22 | 68 |
| 2008 | 13 | 21 | 66 |
| 2012 | 15 | 27 | 58 |
| 2019 | 1 | 47 | 52 |
*net asset value for funds and total assets for other financial intermediaries.
Source of data: Bethlendi – Mérő (2022).
Money market funds belong to monetary financial institutions. They invest in short term liquid assets, typically in government securities, commercial papers, and bank deposits. A specific feature of Hungarian money market funds during the period under review was that, compared to money market funds in advanced economies, the funds' portfolios contained a much smaller proportion of debt securities and a much larger (by several times) proportion of bank deposits, which is held at their banking group's bank. This phenomenon made banks and financial markets more closely interconnected. Figure 2 shows the distribution of the net assets value of Hungarian money market funds between debt securities and bank deposits. In 2017, the EU adopted a new regulation5 effective from 2019, whereby a money market fund may invest a maximum of 10% of its assets in deposit of a given bank. As a result, by 2019, a significant proportion of money market funds have been converted into short term bond funds or have ceased their activity.
Distribution of Hungarian money market funds investments
Source: Eurostat.
Citation: Society and Economy 46, 3; 10.1556/204.2024.00001
The rapid increase in the share of non-money market funds in the last years of the period, as shown in Table 2, was not caused by an increase in equity and/or corporate bond issues, but by the disappearance or conversion of a significant part of money market funds into bond funds. In addition, the asset value of real estate funds also increased significantly in the second half of the 2010s.
The most dominant type of other financial intermediaries is non-bank financial corporations engaged in lending. In Hungary, these FCLs are not engaged in lending combined by securitization, but simple on-balance sheet lending, leasing, factoring, or debt collection. Following the financial crisis, their share declined significantly until 2015, before increasing again. The lending and debt collection activities of other financial intermediaries are not negligible compared to bank lending, but their importance is declining. The share has slightly decreased from around 16% in 2008 to 13% in 2019 (Bethlendi – Mérő 2022).
Despite the underdevelopment of individual shadow banking sub-sectors, the overall size of the Hungarian shadow banking system was not negligible during the period under review shown in Fig. 3. It ranged from 20 to 30% of the banking sector and 14–29% of GDP. It peaked in the third to fourth year after the onset of the financial crisis, both in terms of GDP and banking sector. From 2008 to 2012, new transactions were very limited, but the significant revaluation of foreign currency loans relative to banking sector assets, and the almost stagnant GDP in nominal terms drove the subsequent increase in their proportion.
Balance sheet total of banks and shadow banks to GDP (%)
Source of data: Bethlendi - Mérő (2022).
Citation: Society and Economy 46, 3; 10.1556/204.2024.00001
4.2 Some key shadow banking risks
The traditional balance sheet-based lending activity of the domestic financial system is reflected in the structure of risk-weighted exposures of banks. Looking at the average Pillar I capital requirement composition in 2008–2019, 85% was from credit and counterparty risk, 11% from operational risk and only 4% from market risk (MNB 2020). We do not have such a breakdown for FCLs, but as their main activity is lending, credit risk is also their biggest risk. In addition, liquidity risk is of paramount importance for shadow banking institutions, as they typically finance their activities from volatile sources and are exposed to runs. This is a particular risk for real estate funds, which by definition finance illiquid assets from their volatile funds. We analyze SB risks along the main risk exposures – credit risk and liquidity risk – to demonstrate that even a relatively underdeveloped SB system can pose financial stability risks.
4.2.1 Credit and counterparty risk of non-bank financial companies
We showed that FCLs' activity was one of the most dominant segments of the Hungarian shadow banking system, especially before the 2008 crisis. Because both banks' and FCLs' earnings are significantly driven by credit risk (impairments), we use the volatility of profitability as a proxy for the riskiness of FCLs. Higher volatility of earnings implies higher risk. The standard deviation of the FCLs return on equity (ROE) was significantly higher (24.5%) than that of the banks (13.7%), while their average ROE was almost identical (Fig. 4).
Trends in ROE of banks and FCLs
Source: MNB.
Citation: Society and Economy 46, 3; 10.1556/204.2024.00001
The activity and exposure of FCLs increased significantly after EU accession. The bankruptcy rate6 of FCLs increased substantially during the period hit by the crises (Fig. 5). In the same period, we see far fewer default events in case of commercial banks, demonstrating the relatively higher risk-taking of FCLs. Only one, small commercial bank went bankrupt in 2014 (representing a bankruptcy rate of 2% in that year, in other years it was 0%). By contrast, there was a wave of bankruptcy in the savings cooperative sector in 2015. But we compare FCLs with commercial banks.
Bankruptcy rate of FCLs not belonging to a banking group
Source: Inzelt et al. (2018).
Citation: Society and Economy 46, 3; 10.1556/204.2024.00001
The magnitude of the FCL losses and their contagion in the banking system indicates a significant risk. For several years after 2009, FCLs recorded significant losses, which caused significant losses for the owners and financiers of FCLs. The cumulative losses realized by banks, as owners of FCLs between 2008 and 2014, amounted to nearly 1 billion euros, equivalent to 18% of the post-crisis capital injection by foreign bank owners.7
The potential contagion effect of FCLs is also illustrated by the fact that the MNB (and before 2013 the then separate Hungarian Financial Supervisory Authority, the PSZÁF) required banks to allocate additional capital several times to their FCL portfolio on top of regulatory minimum requirements in the framework of the annual supervisory review process. Some examples are given in Table 3. However, the supervisory capital requirements can only address the risks of the banks providing funds for FCLs. Nevertheless, the risk of FCLs financed from the market, and not part of a banking group, is no less than that of FCLs in banking groups.
Some examples of the PSZÁF/MNB additional capital requirements against banks' exposures to FCLs and other shadow banking entities
| Year | Risk type | Required excess capital |
| 2008 and 2009 | Loans to refinance FCLs granted by banks with own funds less than HUF 1 billion | 0–100% |
| 2011 | Refinancing of the FCLs providing mortgage loans and work-out activities | 50–100% |
| 2014 | Participation in financing of venture capital funds | 150–833% |
Source: MNB archive on supervisory guidance.
4.2.2 Liquidity risk
Shares of public, open-end mutual funds can be redeemed on a daily basis. If the funds' investments are less liquid (e.g. real estate, corporate bonds), then they are exposed to liquidity risk and runs. In Hungary, public open-ended real estate funds have long been widespread. Open-end real estate funds, like commercial banks, provide a liquidity transformation. Their share value is equal to the book-value of the assets minus outstanding debt. They have two types of assets: illiquid real estate investments (the largest part) and liquidity buffer assets. The liquidity risk of real estate is a large concern, especially in down markets (Cheng et al. 2013). The liquidity buffer is a protection against runs (Fecht – Wedow 2014). However, these buffers also represent a trade-off between return and safety. Funds with high liquidity buffers, ceteris paribus, have lower yields, but are safer.
At the end of 2007, prior to the global financial crisis, the stock of Hungarian real estate funds amounted to HUF 498 billion, which accounted for 15% of the domestic mutual funds market. In November 2008, due to the redemption run caused by the loss of confidence in real estate investments, the Hungarian Financial Supervisory Authority had to suspend the trade in real estate mutual funds' shares (PSZÁF 2009). During the suspension period, fund managers had the option to change the fund to a closed-end fund or to change the earlier typical T+3 days settlement period to a maximum T+90 days (Szűcs 2009). By 2015, the net asset value of real estate funds compared to the total net asset value of mutual funds decreased to 10–11%.
The boom of the Hungarian real estate market has caused the volume and share of real estate funds to grow rapidly since 2016. It is now many times higher than before the global financial crisis, with net asset value of HUF 2 290 billion and 33% market share at the end of 2019. Even after the increase of the settlement period from T+3 to a longer one, the typical redemption period remained relatively short in Hungary. Recognizing the related risk, the Hungarian central bank issued a recommendation in 2019 that new public open-ended real estate funds should apply a redemption period of at least 180 days. However, the vast majority of existing funds' shares were issued before the issue of recommendation, so it did not apply to them. Accordingly, the risk of future mass redemption runs is not negligible in the event of a prolonged downturn in the real estate market and a sustained rise in alternative yields (e.g., government bonds). Therefore, in our view, the liquidity risk of real estate funds represents the biggest SB risk at present. The government also recognizes this and issued decree 449/2023. However, it only partly limits this risk.8
5 The secondary shadow banking system
In the early 2000s, before the global financial crisis, the interest rate of the Hungarian forint was much higher than that of the euro or Swiss franc. This paved the way for the mass expansion of foreign currency lending, especially for home mortgages. Foreign currency loans, which were much cheaper than forint loans, were attractive to borrowers and profitable for banks, and the associated exchange rate and interest rate risks were underestimated and ignored by the regulators. The rapid growth in retail foreign currency lending was financed by short-term interbank foreign currency loans from foreign banks and foreign parent banks of Hungarian banks. In the case of Swiss franc loans (which were dominant at that time), they were financed by cross-currency swaps (CCS) linked to parent banks' euro loans (Bethlendi 2015). Interbank lending is not inherently a SB activity (it is part of traditional banking activities) but in this case the funds provided by foreign banks increasingly included market-based sources from SB institutions and activities (Hardie – Howarth 2013; Hardie et al. 2013). Accordingly, the foreign shadow banking funds indirectly became an increasing source of domestic credit growth. The market-based nature of financing has been further reinforced by the extensive use of related derivative instruments. In many cases, banks synthetically generated Swiss franc foreign currency sources (using a combination of parent bank euro loans and cross-currency basis swaps) and used these short-term funds to finance their long-term loans, with a significant rollover risk.
This phenomenon is a well-known and well analyzed feature of the Hungarian banking system. Previous analyses have focused on the role of foreign banks in the rise of foreign currency lending and the materialization of foreign currency lending risk in the years following the 2008 crisis (Banai et al. 2010; Bethlendi 2011; Hudecz 2013; Király 2020). Our approach is substantially different because we do not raise the question of what role foreign banks played in the expansion of FX lending. Rather, we ask to what extent FX lending was financed by shadow bank sources, including domestic and foreign market sources, and how these sources took over the role of bank deposits in banks' funding mix. In other words, we analyze the question of what role (direct and indirect) SB sources played in the pre-crisis boom in foreign currency lending, and how much of a role they have played since the crisis. As presented in the data and methodology section, we analyze banks' funding exposure to SB entities by dividing banks' liabilities into core and non-core liabilities. Non-core liabilities include market-based funds, which we considered to be direct SB financing of banks. The mass foreign interbank loans and CCSs provided to Hungarian banks are defined as secondary SB finance.
The reason why this phenomenon is termed as secondary SB is that banks providing interbank funds to the Hungarian banking system had loan-to-deposit ratios well above 100% in the period under review, especially in the pre-2008 crisis period. For example, the aggregate loan-to-deposit ratios of the German, French and Austrian banks that provided most of the interbank loans to Hungarian banks were respectively 148, 124 and 129% in 2007. Although the ratios for Austria and Germany showed a downward trend, by 2016 only the German banking system had a loan-to-deposit ratio below 100% (Bethlendi – Mérő 2020b). In other words, foreign banks did not pass through deposits (their core liabilities) collected in their home market to Hungarian banks in the form of interbank lending, rather they passed through funds raised from the financial markets, i.e., their non-core shadow banking funds. That is, although the Hungarian SB system is smaller than the SB system in developed European countries, the secondary SB funds among the liabilities of banks may still cause the banking system to be exposed to SB funds.
Figure 6 shows the funding role of market-based liabilities, foreign interbank loans, and central bank funds within the liabilities structure of the Hungarian banking system, where total liabilities are defined as the balance sheet total minus equity. Out of these three types of liability, the pattern of the two non-core liabilities (i.e., the market-based funds and the foreign interbank loans) is significantly different. The share of market-based funds shows a gradual increase, typically up to 1% per year over the period 2004–2015. Then, in 2016, when the earlier depressed bank lending started to recover together with a similar increase in the volume of deposits, the share of market-based funds fell from 20 to 16% in one year. The proportion of foreign interbank loans was still lower than that of the market-based funds in 2004, but a faster growth rate led to an equalization of the share of both funds in banks' liability structure by 2007, followed by the higher share of the latter in the following years. The commitment of the parent banks not to withdraw their funds from their Central-Eastern European subsidiaries during the crisis played an important role in this.
Market-based, foreign interbank, and central bank funding to total liabilities
Source: MNB.
Citation: Society and Economy 46, 3; 10.1556/204.2024.00001
The share of foreign interbank loans peaked at 22% in 2010 and then started to decline (Fig. 6). In 2012, the possibility of Hungarian households to redeem foreign currency loans at favorable rate also significantly reduced the banks’ demand for foreign refinancing of FX loans. In this year, the share of foreign interbank loans in banks' total liabilities declined by more than 5% and has since been on a steady downward trend. This was compounded by the unfolding euro area crisis, which reduced the financing capacity of foreign banks. In addition, Hungarian banking policy also contributed to the decline in the role of foreign funding. The government's banking nationalism, which unfolded in the 2010s, aimed to reduce the role of foreign banks (Johnson – Barnes 2014; Mérő – Piroska 2016; Ban – Bohle 2021). As a result, by 2012, the role of market-based funds had again exceeded that of foreign interbank loans.
Foreign SB funds were also directly present in the liabilities structure of Hungarian banks as foreign market-based funds (Fig. 7), although to a much lesser extent than in the form of secondary sources. The share of market-based funds from the member states of the European Monetary Union ranged between 5 and 10% of total market-based funds of banks over the whole period under review. In other words, foreign SB sources typically financed Hungarian domestic lending secondarily rather than primarily.
Hungarian banks' market-based funds raised from Hungary and from euro area countries (HUF billion)
Source: MNB.
Citation: Society and Economy 46, 3; 10.1556/204.2024.00001
There were several reasons for the decline of foreign interbank funding. First, the global financial crisis led to a significant fall in demand for credit. This resulted in a significant reduction in banks' demand for foreign interbank funding and a steady decline in the loans-to-deposits ratio since 2010. Second, the global financial crisis and euro area sovereign crisis reduced the supply of interbank credit to Hungary. Third, in line with government's so-called self-financing policy, the share of central bank funding to banks under the Funding for Growth Scheme has been steadily increasing since 2013. That is, in addition to the increased funding role of deposits, it was the liabilities provided by the central bank that replaced banks' foreign funding needs and satisfied the banks' demand for funds, which was also in line with the government's financial nationalist banking policy.
In summary, Figs 6 and 7 show that SB funds (both primary and secondary) have appeared among the liabilities of Hungarian banks. In the years preceding the global financial crisis, and thereafter until 2011, the share of secondary shadow banking liabilities in the form of foreign interbank loans and the related cross-currency swaps increased. This only latently increased the SB exposure of domestic banks, as international methodologies for measuring SB, such as those used by the Financial Stability Board and the European Systemic Risk Board (FSB 2015; ESRB 2016), do not take this type of SB presence into account. However, in the case of Hungary, the size of the secondary SB system was comparable to the size of the SB system until 2011, and only became significantly smaller from 2012 onwards (Fig. 8).
SB and secondary SB to GDP (%)
Source: Bethlendi – Mérő (2022) and MNB.
Citation: Society and Economy 46, 3; 10.1556/204.2024.00001
Exposure to secondary SB destabilized the funding model of Hungarian banks at the beginning of 2009 because the market-based renewal of cross-currency swaps linked to interbank lending became impossible. Only the emergency agreement on EUR/CHF swap between Swiss and Hungarian National Banks could have provided renewal possibilities for cross-currency swaps (MNB 2009). As a first step, this led to the materialization of liquidity risk inherent in the foreign currency lending funding scheme, followed by the materialization of credit risk. In other words, our approach to the secondary shadow banking system draws attention to the fact that the materialization of the risks of foreign currency lending in Hungary can also be interpreted as a materialization of the risks arising from the emergence of a secondary shadow banking system.
6 Conclusions
In this article, we have shown that in Hungary, SB activities, which are involved in private sector lending by merging the activities of banks and other financial institutions, are much less developed than in developed countries. Nevertheless, there are financial institutions in Hungary that can be classified as SB institutions. Of these, money market and other investment funds are small in volume by international standards, yet real estate funds have significant liquidity risk. Financial corporations engaged in lending are also present in the Hungarian market, but their activity is based on traditional on-balance sheet lending, rather than securitization. Their activities tend to be riskier than those of banks and given the high proportion of bank refinancing in their liability structure, they are highly interconnected with banks and have a significant potential for contagion to banks.
A specific feature of the Hungarian shadow banking system is what we call secondary SB. This means that market-based funds do not flow to domestic banks through domestic SB institutions, but mainly in the form of foreign interbank loans and the associated cross-currency swaps. In other words, the banking system and the SB activity are not directly interlinked in Hungary, but through the intermediation of foreign banks. The emergence of the secondary SB system made it difficult to recognize the fact that traditional banking had become highly interconnected with SB in Hungary, and the risk of this interconnectedness materialized in the years following the 2008 crisis. Our contribution shows that traditional banking in Hungary was able to attract alternative funding via secondary SB. This feature of banks' funding structure weakens the arguments for the complementary characteristic of the SB system pointed out by the literature (Hodula 2022; Apostoaie – Bilan 2020). The decline of secondary SB, and in general the decline of the involvement of foreign banks in financing the Hungarian banking system in the 2010s, was a result of both economic and political factors.
Through our twofold approach to the Hungarian SB system, we have been able to identify shadow banking risks more accurately than would have been possible through a traditional analysis using an institutional or activity-based approach. We have also shown that the interconnectedness between banking and SB may not only occur through direct exposure, but also indirectly through the presence of secondary SB.
Based on the Hungarian SB case study, a further research question is whether secondary SB is a feature of financial markets in other countries of the CEE region with significant pre-crisis foreign currency lending. And if so, what are the common or specific causes, characteristics, similarities and differences, and implications of this particular path of financial market development? A second further research question is the potential link between the dependent financialisation of the region and the emergence of a secondary shadow banking system in Hungary.
Acknowledgement
The research was supported by National Research, Development and Innovation Office (Grant number: NKFIH138562).
References
Abad, J. – D’Errico, M. – Killeen, N. – Luz, V. – Peltonen, T. – Portes, R. – Urbano, T. (2017): Mapping the Interconnectedness between EU Banks and Shadow Banking Entities. ESRB Working Paper 40.
Acharya, V. – Schnabl, P. – Suarez, G. (2013): Securitization without Risk Transfer. Journal of Financial Economics 107(3): 515–536. http://doi.org/10.1016/j.jfineco.2012.09.004.
Adrian, T. – Ashcraft, A. B. (2012): Shadow Banking Regulation. The Annual Review of Financial Economics 4: 99–140. https://doi.org/10.1146/annurev–financial–110311–101810.
Apostoaie, C. M. – Bilan. I. (2020): Macro Determinants of Shadow Banking in Central and Eastern European Countries. Economic Research–Ekonomska Istraživanja 33. https://doi.org/10.1080/1331677X.2019.1633943.
Bakk-Simon, K. – Borgioli, S. – Giron, C. – Hempell, H. – Maddaloni, A. – Recine, F. – Rosati, S. (2012): Shadow Banking in the Euro Area: An Overview. ECB Occasional Paper 133.
Ban,C. – Bohle, D. (2021): Definancialization, Financil Repression and Policy Continuity in East–Central Europe. Review of International Political Economy 28(4): 874–897. https://doi.org/10.1080/09692290.2020.1799841.
Ban, C. – Gabor, D. (2016): The Political Economy of Shadow Banking. Review of International Political Economy 23(6): 901–914. https://doi.org/10.1080/09692290.2016.1264442.
Banai, Á. – Király, J. – Nagy, M. (2010): The Demise of the Halcyon Days in Hungary: Foreign and Local Banks – Before and After the Crisis. In: The Global Crisis and Financial Intermediation in Emerging Market Economies. BIS Papers 54.
Bethlendi, A. (2011): Policy Measures and Failures on Foreign Currency Household Lending in Central and Eastern Europe. Acta Oeconomica 61(2): 193–223. https://doi.org/10.1556/aoecon.61.2011.2.5.
Bethlendi, A. (2015): Bad Product Development Results in Systemic Market Failure – Foreign Currency Mortgage Loans to Hungarian Households. Financial and Economic Review 14(1): 5–35.
Bethlendi, A. – Mérő, K. (2020a): Árnyékbankrendszer a közép–és kelet–Európai régióban [Shadow Banking in the Central and Eastern European Region] Statisztikai Szemle 98(10): 1151–1184.
Bethlendi, A. – Mérő, K. (2020b): Changes in the Structure of Financial Intermediation – Eastern–Central European Developments on the Light of Global and European Trends. Danube 11(4): 283–199. https://doi.org/10.2478/danb–2020–0017.
Bethlendi, A. – Mérő, K. (2022): Measuring Shadow Banking in Central and Eastern European Countries: A New Dataset, 2004–2019. Regional Statistics 12(4): 112–131. https://doi.org/10.15196/RS120405.
Cheng, P. – Lin, Z. – Liu, Y. (2013): Liquidity Risk of Private Assets: Evidence from Real Estate Markets. Financial Review 48(4): 671–696.
Claessens, S. – Ratnovski, L. (2014): What Is Shadow Banking? IMF Working Paper 14/25.
Dooly, M. – Hutchison, M. (2009): Transmission of the U.S. Subprime Crisis to Emerging Markets: Evidence on the Decoupling–Recoupling Hypothesis. Journal of International Money and Finance 28(8): 1330–1349. https://doi.org/10.1016/j.jimonfin.2009.08.004.
ESRB (2016): EU Shadow Banking Monitor No 1. https://www.esrb.europa.eu/pub/pdf/reports/nbfi_monitor/20160727_shadow_banking_report.en.pdf.
Fecht, F. – Wedow, M. (2014): The Dark and the Bright Side of Liquidity Risks: Evidence from Open–End Real Estate Funds in Germany. Journal of Financial Intermediation 23(3). https://doi.org/10.1016/j.jfi.2014.02.002.
FSB (2011): Shadow Banking: Scoping the Issues. A Background Note of the Financial Stability Board. https://www.fsb.org/wp–content/uploads/r_110412a.pdf.
FSB (2020): Global Monitoring Report on Non–Bank Financial Intermediation. https://www.fsb.org/2020/12/global–monitoring–report–on–non–bank–financial–intermediation–2020/.
FSB (2015): Global Shadow Banking Monitoring Report .https://www.fsb.org/wp–content/uploads/global–shadow–banking–monitoring–report–2015.pdf.
Ghosch, S. – Gonzalez del Mazo, I. – Ötker–Robe, I. (2012): Chasing the Shadows: How Significant Is Shadow Banking in Emerging Markets? World Bank Economic Premise 88.
Giron, C. – Matas, A. (2017): Interconnectedness of Shadow Banks in the Euro Area. IFC– National Bank of Belgium Workshop, BIS Irving Fisher Committee on Central Bank Statistics, https://www.bis.org/ifc/publ/ifcb46r.pdf.
Hodula, M. (2022): Bringing the Flashlight: Shadow Banking in European Union Countries. Finance Research Letters 47: 102688. https://doi.org/10.1016/j.frl.2021.102668.
Hodula, M. – Melecky, A. – Machacek, M. (2020): Off the Radar: Factors behind the Growth of Shadow Banking in Europe. Economic Systems 44(3): 100808. https://doi.org/10.1016/j.ecosys.2020.100808.
Harutyunyan, A. – Massara, A. – Ugazio, G. – Amidzic, G. – Walton, R. (2015): Shedding Light on Shadow Banking. IMF Working Paper WP/15/1.
Hardie, I. – Howarth, D. – Maxfield, S. – Verdun, A. (2013): Banks and the False Dichotomy in the Comparative Political Economy of Finance. World Politics 65: 691–728. https://doi.org/10.1017/S0043887113000221.
Hardie, I. – Howarth, D. (2013): Market–Based Banking and the International Financial Crisis. Oxford: Oxford University Press.
Hudecz, A. (2013): FX Lending to Households in Poland, Romania and Hungary, 1997–2011. Acta Oeconomica 63(3): 257–286. https://doi.org/10.1556/aoecon.63.2013.3.1.
Inzelt, G. – Szentes–Markhot, Z. – Budai, G. (2018): Monitoring of Banks’ Risks Related to the Funding of Financial Enterprises. Financial and Economic Review 17(4): 112–139.
Johnson, J. E. – Rarnes, A. (2014): Financial Nationalism and Its International Enables: The Hungarian Experience. Review of International Political Economy 22(3): 535–569. https://doi.org/10.1080/09692290.2014.919336.
Király, J. – Mérő, K. (2010): No Free Lunch – No Decoupling, the Crisis and Hungary: A Case Study. In: Gup, B. E. (ed.) The Financial and Economic Crises – An International Perspective. Cheltenham UK, Northampton, MA, USA: Edward Elgar, pp: 79–102.
Király, J. (2020): Hungary and Other Emerging EU Countries in the Financial Storm. Cham: Springer.
Kjosevski, J. – Petkovski, M. – Stojkov, A. (2021): What Drives Shadow Banking in the New EU Member States? Empirical Panel Cointegration Approach. Ekonomicky Casopis 69(8): 812–832. https://doi.org/10.31577/ekoncas.2021.08.03.
Koroknai, P. – Lénárt-Odorán, R. (2011): The Role of Special Purpose Entities in the Hungarian Economy and Statistics. MNB Bulletin, October. https://www.mnb.hu/letoltes/koroknai–lenartodoran–en.pdf.
Lysandrou, P. – Nesvetailova, A. (2015): The Role of Shadow Banking Entities in the Financial Crisis: A Disaggregated View. Review of International Political Economy 22(2): 257–279. https://doi.org/10.1080/09692290.2014.896269.
Malatesta, F. – Masciantonio, S. – Zaghini, A. (2016): The Shadow Banking System in the Euro Area: Definitions, Key Features and the Funding of Firms. Italian Economic Journal 2(2): 217–237. https://doi.org/10.1007/s40797–016–0032–0.
Mérő, K. – Piroska, D. (2016): Banking Union and Banking Nationalism – Explaining Opt–Out Choices of Hungary, Poland and the Czech Republic. Policy and Society 35(3): 215–226 https://doi.org/10.1016/j.polsoc.2016.10.001.
MNB (2009): Swiss National Bank and Magyar Nemzeti Bank Cooperate to Provide Swiss Franc Liquidity. Press Release. https://www.mnb.hu/en/pressroom/press–releases/press–releases–2009/swiss–national–bank–and–magyar–nemzeti–bank–cooperate–to–provide–swiss–franc–liquidity.
MNB (2015a): Financial Stability Report (May). https://www.mnb.hu/en/publications/reports/financial–stability–report/financial–stability–report–may–2015.
MNB (2015b): Financial Stability Report (November). https://www.mnb.hu/en/publications/reports/financial–stability–report/financial–stability–report–november–2015.
MNB (2020): Macroprudential Indicators of Financial Stability Report (May). https://www.mnb.hu/en/publications/reports/financial–stability–report/financial–stability–report–may–2020.
Pires, F. (2019): Non-banks in the EU: Ensuring a Smooth Transition to a Capital Markets Union. SUERF Policy Note 103.
Pozsar, Z. – Adrian, T. – Ashcarft, A. – Boesky, H. (2010): Shadow Banking. Federal Reserve Bank of New York Staff Report. 458.
PSZÁF (2009): Éves jelentés [Annual Report] 2008, https://www.mnb.hu/letoltes/pszaf–eves–2008–nyomtatasra.pdf.
Raiffeisen (2013): A Raiffeisen Bankcsoport kockázatkezelésre vonatkozó információinak nyilvánosságra hozatala [Risk Management Disclosure Report of Hungarian Raiffeisen Group] https://www.raiffeisen.hu/documents/bank/jelentes/eves/kockazatkezelesi_informaciok_bankcsoport_2013.pdf.
Shin, H. S. – Shin, S. (2011): Prociclicality and Monetary Aggregates, NBER Working Paper Series 16836.
Szücs, J. (2009): Az ingatlanalapok jogi szabályozása és gazdasági helyzete [Regulation and Economic Situation of Real Estate Funds]. Hitelintézeti Szemle 8(6): 528–555.
Thiemman, M. (2018): The Growth of Shadow Banking. Cambridge: Cambridge Books.
Thiemann, M. – Tröger, T. H. (2021): Detecting Tail Risks to Preclude Regulatory Arbitrage. The Case for a Normatively Charged Approach to Regulating Shadow Banking. Accounting, Economics, and Law: A Convivium 11(2): 233–270. https://doi.org/10.1515/ael–2019–0059.
Although insurance companies and pension funds are not institutionally considered to be part of the shadow banking system, the funds they provide are non-core funds, as they are backed by market funds raised from investors.
For example, in the early 2010s, Hungarian Raiffeisen Bank's parent company guaranteed the mezzanine tranche of a loan portfolio securitized by the Hungarian bank, while the senior and junior tranches remained on the bank's balance sheet, i.e., no market sale took place. A detailed description of such a transaction can be found, in Hungarian Raiffeisen Bank's 2013 disclosure report (Raiffeisen 2013: 81–85).
The practice of using money market or bond funds to provide funding to banks in the form of short-term repo transactions.
Repos, as deposit substitutes, are part of the broader money supply (M3) and are included in the monetary statistics.
Regulation 2017/1131. According to this regulation, money market funds must comply with much stricter investment, risk management, and disclosure rules.
Bankruptcy is defined as the number of liquidation, bankruptcy proceedings, removal by the court, completed liquidation, forced dissolution within one year after the balance sheet date of the respective annual report.
Authors' estimation based on MNB (2015a, 2015b).
In case of public, open-end real estate funds the share of the stock of investment unit series with at least T+180 days redemption should exceed 15% of the total value of the assets that qualify as real estate by 2026.
