THE IMPORTANCE OF THE BANK TAX FOR THE POLISH STATE BUDGET − HAS THE COVID-19 PANDEMIC REDUCED THE PROCEEDS FROM THIS TAX?

: The article’s main objective was to analyse the implications of the tax on certain financial institutions, referred to as the bank tax, for the Polish state budget since the very beginning of the levy. The article also indicates the essence and the purpose of this tax’s functioning, in comparison to the European Union countries, as well as presents the principles of its legal structure. Based on the research carried out, it was found that the level of the bank tax contributions to the state budget in 2016-2019 was stable, assuming a slight upward trend that continued in 2020, despite the outbreak of the COVID-19 pandemic. The Polish government’s additional financial activities, induced by the pandemic, have increased the tax base in the bank levy. These activities have also contributed to the strengthening of the links between the government and the banking sectors, measured by the sovereign-bank nexus index, as a result of the continued increase in the banks’ exposure to Treasury securities. These securities are excluded from the bank-levy tax base. The following research method was used in the preparation of the article: a critical study of the literature and an analysis of the empirical data published by the Polish Financial Supervision Authority, the Ministry of Finance, the Supreme Audit Office, and the National Bank of Poland.


Introduction
The bank tax, officially understood as the tax on certain financial institutions, has been present in the Polish public finance system for over five years. It seems that after this time, it is possible to assess its significance for the state budget revenues in greater detail. Moreover, the COVID-19 pandemic, which appeared in Poland in March 2020, has completely changed the conditions of many entities' functioning in the economy (cf. Giżyński, 2021, pp. 33-34). As a result of the economic slowdown (NBP, 2021a, p. 13) (in 2020, real GDP decreased by 2.7% in Poland) (Rada Ministrów, 2021, p. 61) and the anti-crisis measures taken by the Polish government (according to official data, their scale exceeded 4% of GDP in 2020), aimed at mitigating the effects of the pandemic, the public finance sector's condition has deteriorated significantly (NBP, 2021a, p. 13). The state budget tax revenues, which include the 2020 proceeds from the bank levy, experienced a nominal increase of only 0.8% year on year. Bank tax revenues, in turn, increased in nominal terms by 2.6% year on year (Rada Ministrów, 2021, p. 63). The nominal increase in the bank levy revenues resulted from the fact that the factors increasing the tax base were predominant in the banking sector's assets, despite the increase in the value of the Treasury securities lowering this base (Rada Ministrów, 2021, p. 80). The 2020 increase in the banks' exposure to Treasury securities and Treasury guaranteed bonds by nearly 4 pp in the banking sector's assets (NBP, 2021a, p. 73) has contributed, among others, to the increased links between the government and the banking sectors. The strength of these links can be measured by the sovereignbank nexus index (NBP, 2020b, pp. 7, 73;2021a, pp. 72-73). In 2020, this index approached its maximum (NBP, 2021b).
The main objective assumed in the article was to analyse the implications of the tax on certain financial institutions for the Polish state budget since the very beginning of the levy. The introductory part describes the theoretical and the systemic aspects of the bank levy. The research period was divided into two phases; the first, covering the years 2016-2019, involved detailed analysis of the way the state budget's tax revenues were shaped in those years, including a discussion regarding the increase in the banks' demand for Treasury securities as a consequence of the adopted legal structure of the tax, as well as and an assessment of the tax collection supervision. The second phase, concerning 2020, was discussed in this study in the context of the COVID-19 pandemic. The main research hypothesis was formulated as follows: the upward trend in the revenues from the bank tax to the Polish state budget has continued, despite the outbreak of the COVID-19 pandemic.
The research method used in the course of the article's preparation involved a critical study of the literature as well as analyses of the empirical data published by the most important public (supervisory) institutions in Poland.

Bank levy -the essence and purpose of its functioning in European Union countries
The financial crisis of 2007 showed that the banking institutions' engagement in risky operations can have dramatic consequences. In many European Union countries, the cost of saving failing banks entailed an increase in public debt, and as a consequence, some of these banks faced bankruptcy. In order to reduce the risk of further crises, international organizations and financial market supervisors decided to reform the banking sector. As part of the discussion on this reform, the International Monetary Fund (IMF) proposed the introduction of a bank levy. Its main purpose was to reduce banks' engagement in risky transactions (in the case of bank asset taxation), as well as to limit their exposure to risky financing obtained from the interbank market (tax on liabilities). The introduction of the bank tax, therefore, was meant to reduce both the risk in the banking sector and the likelihood of such potential systemic crises as those experienced in 2008-2010 (Hryckiewicz, Mielus, Skorulska, & Snarska, 2018, p. 2). According to the general definition, the bank tax is a public law burden directed at a closed group of financial entities, i.e. banks and some shadow banks. Broadly speaking, it encompasses the taxes levied on operations or certain revenues, and also applies to bank charges. Accordingly, it may take the form of an indirect or a direct tax (Giżyński, 2017, p. 30).
On the European Union (EU) forum, the discussion regarding the implementation of a tax levied on financial institutions began in 2010. From the very beginning it mainly concerned a tax on financial transactions. Subsequent analyses, however, indicated the potential low effectiveness of introducing this type of levy, together with the possibility of moving the activity of a part of the EU financial sector outside the EU borders, namely to other global financial centres. Therefore, despite the initial interest in this concept (including the largest EU economies, Germany and France), the decision was made not to adopt a uniform bank tax within the EU community (Martysz & Bartlewski, 2018, pp. 111-112).
Most of the EU countries introduced the bank levy as compensation for the public aid granted for the restructuring of financial institutions during the financial crisis 1 . These countries were: Austria, Belgium, Cyprus, France, the Netherlands, Latvia, Germany, Portugal, Slovenia, Sweden and the United Kingdom. Nevertheless, this tax was also introduced where the above-mentioned aid was not granted, in such countries as Finland, Romania, Slovakia and Hungary. Poland is also one of those countries (KNF, 2020, p. 15). Based on the information in Table 1, it can be noted that in most countries the bank tax revenue is assigned to the central budget. Germany, Romania, Slovenia and Sweden, where this tax constitutes the source of financing special restructuring funds, are an exception here.
In most EU member states the bank tax was introduced with regard to balance sheet values (see Table 1). The objectives behind the introduction of this levy include primarily the fiscal objective, as well as the improvement of the state's financial stability and accumulation of appropriate funds. These funds are to, as already emphasized, cover the expenditures incurred for the repair of the financial sector,  (Martysz & Bartlewski, 2018, p. 113). B In 2016, a new tax system was adopted. Since then, part of the funds goes to the Belgian Settlement Fund, and part to the Common (European) Settlement Fund (for more, see: EC, 2020). C This tax was introduced temporarily, i.e. for the years 2013-2015 (Kozłowska, 2017, p. 90). D Since 2019, the banks in Romania pay an additional tax on assets (Stellmaszyk, 2019). E As of the 1 st of July 2020, the tax was abolished (Szewska, 2020). F In 2016, payments to this fund were replaced by the resolution fee, which then began to supply the new resolution reserve (some funds from the Stability Fund were transferred there as well). The Stability Fund, however, remained in force, and part of its resources was intended to serve as preventive funding under the Debt Office (The Swedish National..., 2020).
Source: own elaboration based on (Autoridade Tributaria..., 2020;Bremus, Schmidt & Tonzer, 2020, p. 16;Buch, Hilberg & Tonzer, 2016, p. 54;Deloitte, 2020;EC, 2020;IMF, 2018, p. 39;Jarno & Kołodziejczyk, 2018, pp. 88-89;Martysz & Bartlewski, 2018, p. 113;OECD, 2020;PWC, 2020;Santander, 2020; Zákon o osobitnom...). or those to be possibly incurred in the future in that regard (KNF, 2020, p. 15). The most common tax base that the member states selected for their bank tax was the liabilities reduced by equity. As Table 1 shows, several countries chose a different base, namely Finland, France, Hungary and Slovenia; Poland is also in this group. In Poland, the bank tax was introduced with a significant delay when compared with most EU countries (see Table 1). More than seven years after the bankruptcy of Lehman Brothers, the arguments for the imposition of an anti-crisis levy did not fit into the reality of over a 2.5% GDP growth. One of the factors that supported the imposition of the above burden was the change in public sentiment. A debate on the operating conditions of the banks in Poland and their low fiscal burden was ongoing at the time. This burden was considered disproportionate to the scale of profits achieved by these institutions. These arguments determined the introduction of additional taxation for the banking institutions in Poland (Giżyński, 2021, pp. 35--36;Graca, 2020, p. 55).

Principles of the bank tax in Poland -with commentary
Poland's Banking Tax Act was processed very quickly and adopted on 15 January 2016, exactly 43 days after the parliamentary bill had been submitted to the lower house of the Parliament -the Sejm. The fast pace of the legislative work resulted from the need to obtain an additional source of financing for budgetary expenses. This referred to the social expenses in particular (NIK, 2018, pp. 5, 8), as the implementation of an election promise from the autumn 2015 parliamentary election campaign (Giżyński, 2017, p. 32). The introduction of the tax was additionally justified by the wish to increase the financial sector's share in state expenditure financing (NIK, 2018, p. 5). This means that in Poland, the adoption of an additional bank levy was not related to any specific financial-system risks nor to the public aid previously granted to the financial sector. The Polish bank tax was exclusively of a fiscal nature (Giżyński, 2017, p. 33).
It should be emphasized that the legal provisions regarding the bank tax in Poland were included in the Act of 15 January 2016 on the tax on certain financial institutions (Ustawa z dnia 15 stycznia 2016 r.). These provisions came into force on February 1, 2016. It is worth adding that the term 'bank tax' is colloquial. In addition to banks, this levy is paid, among others, by insurance companies and loan institutions (see Table 2). The simplified term of 'bank tax' seems to be justified (Martysz & Bartlewski, 2018, p. 116), since over 85% of this tax is paid by the largest banks (Giżyński, 2021, p. 39).
The entire amount of the tax on certain financial institutions contributes to the income of the state budget. The subject of this tax are the assets of the institutions that are bank levy taxpayers (Ustawa z dnia 15 stycznia 2016 r., art. 2 & 3). The taxable base is the excess of the total value of the taxpayer's assets (see Table 2). The surplus is determined on the basis of the taxed entity's statement of turnover and balances on the last day of the month, based on the entries from its general ledger accounts. The tax base can be reduced (see Table 2), among others, by the value of the assets constituting Treasury securities (TS) 2 or the funds accumulated under the contracts for management of employee capital plans (Pracownicze Plany Kapitałowe, PPK) (Ustawa z dnia 15 stycznia 2016 r., art. 5). The monthly bank tax rate in Poland is set at 0.0366% of the tax base (Ustawa z dnia 15 stycznia 2016 r., art. 7). The taxpayers are obliged to complete their own tax declarations using the FIN-1 template, then submit the declarations to the head of the competent tax office, as well as calculate and pay the tax amount due by the 25 th day of the month following the month which this amount pertains to. It should be noted that tax declarations are only submitted by entities that are to pay the tax (Ustawa z dnia 15 stycznia 2016 r., art. 8). The first bank tax settlement period was February 2016 (Ustawa z dnia 15 stycznia 2016 r., art. 13). The legislator also provided for some reliefs and exemptions in the bank levy. State-owned banks have been exempted from it (Ustawa z dnia 15 stycznia 2016 r., art. 10 & 11); so far 3 , the only such bank in Poland, within the meaning of the law, is the Bank Gospodarstwa Krajowego (BGK). Apart from the state-owned banks, entities that are in liquidation, suspension or bankruptcy, as well as those covered by the recovery program, have also been exempted from paying the bank tax (Giżyński, 2017, pp. 32-36;NIK, 2018, p. 5). The final part of the Banking Tax Act contains a provision stating that the imposition of the new levy on certain financial institutions may not be the grounds for changing the terms of financial and insurance service provision. This referred to the contracts concluded before the effective date of this act (Ustawa z dnia 15 stycznia 2016 r., art. 14). The above-mentioned provision was to prevent the phenomenon of bank levy shifting 4 (Giżyński, 2017, p. 36).
At the initial stage of the bank levy's operation, the Ministry of Finance did not classify it as being particularly vulnerable to irregularities. When assessing this risk, the fact that banking and insurance institutions enjoy the status of public trust was taken into account. This decision was also influenced by the method of determining the tax base and calculating the tax, as well as the method of tax accounting. What is more, it was found that the risk of irregularities in the settlement of the bank tax is acceptable, while the risk of avoiding audit or the risk of non-cooperation between the taxpayers and the audit authorities is low (NIK, 2018, p. 12).
Analyses of the bank tax legal structures adopted in European countries show that the bank levy rules were adapted to the objectives these countries wanted to achieve. The first objective was to obtain additional income for the central budget, and the second was to maintain the stability of a given country's financial sector (Majchrzycka--Guzowska, 2019, p. 332). Against the background of the EU solutions, the Polish bank tax constitutes a necessary levy. Nevertheless, it has been the subject of criticism since the very beginning, because it is based on assets, while its rate is many times higher than in other Member States. It should be emphasized that the adopted tax--collection objective is redistributive in character. Such an objective is justified from the state budget revenue perspective (Martysz & Bartlewski, 2018, p. 118).

Analysis of the proceeds from the bank levy
Based on the first four years (2016-2019) of the operation of the bank tax in Poland, it can be observed (see Table 3), that the level of tax payments to the state budget was stable, assuming a slight upward trend -the mid-term pace of changes amounted to 7.24%. The lower value of inflows in 2016 resulted from the fact that the tax collection period was shorter by two months -the first payments made by financial institutions were settled in March 2016 (see Table 3). In the period analysed, the total revenues from this levy amounted to approximately PLN 17 056 million.  Ministrów, 2017Ministrów, , p. 46, 2018Ministrów, , p. 45, 2019Ministrów, , p. 46, 2020).
The upward trend is justified by the increase in the banking sector's assets, which constitute the basis for calculating the bank tax. Moreover, at the turn of 2018 and 2019, a significant increase in the revenues from this tax occurred. The likely cause of this increase entailed the process of consolidation in the Polish banking sector 5 . This process accelerated starting from mid-2018 (Czechowska, Hajdys, Stawska, Zatoń, & Sikorski, 2020, p. 22).
During the analysed period, the effectiveness of bank tax revenue planning in Poland fluctuated. In the first year (2016) of the levy's validity, its revenue plan was achieved only in 63.8% (see Table 3). The inflated forecast resulted from the fast pace of the work on the adoption of the act regulating this levy. When the 2016 budget was prepared, the final version of the bill was not yet known. As already mentioned, the government planned the bank tax revenues based on a private members' bill. Its content differed from the final version, however. It is worth adding that the Ministry of Finance representatives explained these discrepancies by the fact that they occurred during the introduction of sectoral taxes. Moreover, it was argued that such discrepancies occurred in other countries that decided to introduce a bank tax (for more, see: NIK, 2018, pp. 15-16).
The bank levy planning efficiency improved significantly in 2017. This time the actual revenues were 10.3% higher than planned (see Table 3). This result can be justified by the fact that the revenue forecast was estimated based on the data from the first four months of the levy collection in 2016. The higher revenues primarily resulted from a faster than assumed increase in the value of the assets taxed. The amounts that were supposed to reduce the tax base were ultimately lower (Rada Ministrów, 2018, p. 45).
In the years 2018-2019, the differences between the planned and the realized bank tax revenues were much smaller. In 2018, the final proceeds were only 1.3% lower than planned (see Table 3). This was mainly due to a slower than assumed increase in the value of assets, with higher amounts of both non-taxable assets and those reducing the tax base (Rada Ministrów, 2019, p. 46). The situation reversed in 2019, when the revenues realized were 3.3% higher than forecasted (see Table 3). A higher increase was noted in the value of the taxable institutions' average monthly assets, i.e. by 7.1%. At the same time, the average monthly amounts reducing the tax base increased by 10.5% -nominally, this increase was much smaller than the increase in the value of assets (Rada Ministrów, 2020, p. 47).
With the exception of 2017, the 2016-2019 share of the bank tax in the state budget tax revenue was at a similar level (see Table 4). The average value of this indicator was 1.31% at the time. The shares of the bank levy in the total budget revenues did not differ significantly in individual years either. The lowest share was 1.11% in 2016, whereas the highest was 1.24% in 2017. The average for this indicator, during the period analysed, was 1.18%. Then again, the bank tax's share in nominal GDP turned out to be the least diversified; its average value for the period 2016-2019 was 0.21%. It should be noted that the values of each of the above three indicators were the lowest in 2016 (for more, see Table 4), whereas the tax collection period, which was shorter by two months, should be considered the main reason for this. Source: own elaboration based on (Rada Ministrów, 2017Ministrów, , pp. 9, 34, 2018Ministrów, , pp. 9, 32, 2019Ministrów, , pp. 9, 34, 2020. The dominant share in the tax on certain financial institutions was that of the banking sector 6 , slightly over 85% on average (see Table 5). Most of this tax was paid by the largest banks operating in Poland. The share of nine such entities in the bank levy payments was, on average, close to 80% in 2016-2019 (see Giżyński, 2021, p. 39). Domestic insurance and loan companies paid a total of less than 15% (see Table 5). Due to the non-achievement of an asset value in the amount of PLN 4 billion, and as a result of other possibilities of tax base reduction, including exemptions, the tax on certain financial institutions was not paid by credit unions or cooperative banks (Czechowska, et al. 2020, p. 22). Source: own elaboration based on (KNF, 2020, p. 8;NIK, 2018NIK, , p. 19, 2019NIK, , p. 97, 2020Rada Ministrów, 2017, p. 46, 2018, p. 45, 2019, p. 46, 2020).
The bank tax arrears varied in the period under analysis. The tax collection audit carried out by the Supreme Audit Office with regard to the period of February 2016 − September 2017 showed that, in the tax offices audited, these arrears mainly resulted from the audit activities carried out at the time. These activities pertained to the corporate income tax overpayments and their reclassification, on the taxpayer's request, as the bank tax (NIK, 2018, pp. 22, 40). At the end of 2019, the State Treasury's receivables from this tax totalled approximately PLN 2.5 million (Rada Ministrów, 2020, p. 334).
The imposition of the bank tax had the intended effect, taking the form of high revenues, which constituted slightly more than 1% of the state budget income. Financial institutions, mainly the largest banks in Poland, paid a total of approximately PLN 4.5 billion a year in this tax (see Table 3). As per the legislator's intention, the funds obtained in this way could help finance the state's new social programs. The question is, however, how long such amounts will be obtained. Even in times of prosperity, domestic banks had difficulties with ensuring profitability of their capital 7 (Graca, 2020, pp. 54, 65-66).

Increase in the banks' demand for Treasury securities
Apart from the increasing state budget revenues, another aftereffect associated with the introduction of the tax on certain financial institutions entailed an increase in the banks' demand for Treasury securities (see Figure 1). These institutions shifted their activity from the interbank market to the Treasury securities market (Martysz & Bartlewski, 2018, p. 125). Only in January and February 2016, domestic banks increased their exposure to these securities by almost PLN 35 billion (NBP, 2016, p. 27). By investing in Treasury bonds, banks reduce the bank levy tax base and, consequently, the tax itself. Although investments in Treasury securities bring a low rate of return, they are more advantageous for these entities, as opposed to allocating funds to a non-returnable tribute. Such behaviour on the part of the banks also benefits the government, which finances the budget expenses via the sale of Treasury securities (Graca, 2020, pp. 59-60). What is more, this also makes state debt servicing easier (Giżyński, 2017, p. 38).
In 2016-2019, the value of the Treasury securities in banks' portfolios increased annually by 10.95% on average, whereas in total, it increased by PLN 139.9 billion, 0.93 pp on average, had it not been for the bank levy. The highest difference was 0.96 pp in December 2016, whereas the lowest was 0.89 pp in December 2018. In 2019, however, the bank tax reduced ROE by 0.93 pp (KNF, 2020, p. 18).
i.e. by 72.4% (see Figure 1). In 2019, Treasury securities accounted for 71.8% of the value of the banks' entire securities portfolio and 16.7% of the value of the banking sector's assets. In the anaylsed period, both indicators increased by 8.9 pp and 1.5 pp, respectively (NBP, 2020c, p. 106).
In addition, the Supreme Audit Office auditors 8 found that domestic banks did not use Treasury security operations for tax optimization. In turn, an optimization mechanism, entailing an increase in the purchase of Treasury securities in the last days of the month and their sale in the first days of the following month, was observed in credit institution branches. Due to the low 2% share of the Treasury securities held by these institutions, the auditors assessed that these transactions did not significantly affect the amount of the bank tax. The domestic banks' share in these securities was approximately 98% (NIK, 2018, pp. 10-11).
Experts predict that banks will continue to increase their Treasury security purchases (Raport o sytuacji..., 2020, p. 256). The bank tax relief creates incentives for these institutions to increase their exposure to the government sector. Nevertheless, the complexity of the relationship between the two sectors is deepening. It should be remembered that as a result of the banks' purchases of Treasury securities, the government sector, namely the Ministry of Finance, remains both the banking sector's regulator and its debtor. Moreover, in recent years the involvement of the public authorities in the largest Polish banks has increased. This increase is manifested in the dominant ownership share, with concurrent significant representation in the financial supervision decision-making bodies. When the government is simultaneously the owner, the creditor, and the supervisory authority of systemically crucial banks, a number of challenges to the financial stability result (for more, see: NBP, 2019a, pp. 124-127). This is due to the fact that strong dependencies between the government and the banking sectors may result in negative feedback in connection with their financial situation (NBP, 2019b, pp. 122-123). A bank's majority owner may be subject to market pressure regarding the entity's recapitalization when it experiences financial difficulties, for example as a result of tensions on the debt market (NBP, 2019a, p. 126).
The complexity degree of the banks-authorities relationship is determined by the sovereign-bank nexus index (NBP, 2019b, pp. 122-123), which can be calculated as a product of percentile ranks of the following three variables constituting it: 1) regulatory capital's vulnerability to the credit risk of Treasury Bonds; 2) share of the government representatives' votes in the KNF votes; 3) share of the banks controlled by the State Treasury in the sector's assets. Due to the structure of the sovereign-bank nexus index, its values range from 0 to 1. An index value approaching 1 indicates that the levels of its three components are close to the highest values (NBP, 2021a, p. 73). The December 2014-December 2019 development of this index, including the variables forming it, is shown in Figure 2. As can be seen, a significant increase in, among others, the variable describing the sensitivity of the banking sector's equity to the credit risk associated with Treasury bonds occurred in 2016. At the end of 2015, this variable's value was 0.03; after the first quarter of 2016, it was 0.08, whereas at the end of 2016, it was 0.15. The entire index increased by 0.39 points in 2016, i.e. up to 0.46. In 2017-2019, on the other hand, the value of the variable describing the sensitivity of equity to the credit risk associated with Treasury bonds was at the level of 0.16 on average, deviating, at the same time, by 0.04 points. At the end of 2019, this variable was 0.18. The entire sovereign-bank nexus index also increased slightly, by 0.02 points, up to 0.48 (see NBP, 2021b). The global financial crisis of 2007-2009, as well as the eurozone's debt crisis observed in 2010-2012 in some EU member states (Giżyński & Wierzba, 2018, p. 186), revealed a strong link between the public finance sector's condition and that of the banking sector. A shock transmission mechanism could at the time be observed in Ireland, Greece, Portugal and Spain. This mechanism is bi-directional, and mainly results from banks' exposure to public debt. A decline in a state's creditworthiness has an adverse impact on the value of the Treasury bonds in banks' portfolios. This causes domestic banks to incur losses and increases the probability of these banks being granted public aid, which exerts pressure on the state budget. Furthermore, banks with the highest losses reduce their lending, which translates into a decline in economic activity and weakens the borrowers' financial condition. Tax revenues decrease as a consequence (NBP, 2019a, p. 126).
The legal structure of the bank tax in Poland should include such solutions as to ensure increasingly higher revenues to the state budget, making the methods of tax evasion, such as increasing the purchases of Treasury bonds, unprofitable, while maintaining a safe level of the banking system risk. Furthermore, such solutions are intended to encourage banks to grant loans and create value in the economy. As such, it is worth considering a tax base change in the levy on certain financial institutions, from assets to profit, or the establishment of its calculation on the basis of the average levels of the balance sheet items (Martysz & Bartlewski, 2018, p. 130).

Supervision of bank tax collection and an assessment of its effectiveness
The Ministry of Finance took steps to prevent bank tax evasion. Immediately after the introduction of the new levy, a general interpretation was issued, aimed at the uniform application of the provisions contained in the Act. Moreover, in 2016 the interpretation was presented three times with regard to the possibility of reducing the tax base by the value of the securities not covered by the regulations. In one case, the officials decided that the securities issued by the European Investment Bank (EIB) 9 would be treated in Poland as Treasury securities (NIK, 2018, p. 11).
Additionally, as a result of the position taken by the Ministry of Finance, several taxpayers submitted corrections to their tax returns. The resultant situation increased their bank tax liabilities for the period February-May 2016 by a total of PLN 3.4 million (NIK, 2018, pp. 34-35).
A situation emerged where the individual interpretations issued by the tax authorities in 2018 were changed in favour of the taxpayers, after the first-instance judgments of the administrative courts in this regard had become final (Dudek, 2020, p. B5). However, during the first three years of the bank tax collection the tax authorities interpreted the provisions to the disadvantage of the taxpayers, especially with regard to insurance companies. The auditors argued that the amount exempted must be shared between the capital group companies that are controlled by both Polish and foreign entities. As a result, foreign entities were settling the bank levy on unfavorable terms (Dudek, 2019b, p. B2). Two insurance companies whose majority shares are held by foreign companies were in a dispute with the tax administration over this matter (Dudek, 2019a, p. B2). The court of cassation has repeatedly ruled that the provision regarding the joint exempt amount applies to subsidiaries only. It indicated that no dependency occurred in the cases examined, where the control was exercised by a foreign company, which does constitute a link, but in a broader sense. It was additionally argued that the tax authorities, appointed among others to interpret the law, cannot seal the system via changes in the tax regulations, since this activity is reserved for the legislator exclusively (Dudek, 2019b, p. B2). The cassation court also emphasized that the authorities' arguments addressing the purpose of the act, the unclear language as well as the legislator's rationality, were not sufficient enough as it is the actual wording of the provisions that is of key significance. Subsequent judgments were quickly issued after the cassation court's rulings. The courts of first instance began to adjudicate in the same tone (for more, see : Dudek, 2019a: Dudek, , p. B2, 2019b. At the same time, the tax authorities stopped submitting cassation appeals to these rulings, while the Director of the National Revenue Information changed the interpretations contested (Dudek, 2020, p. B5).
The Supreme Audit Office positively assessed the Minister of Finance's supervision over the bank levy collection in the first 20 months of its operation. Nevertheless, it indicated that, at the level of all tax offices, this supervision lacked a uniform tax management mechanism. It was explained that such a good-practice--based mechanism could be used to inspect the entities that potentially were subject to taxation (for more, see: NIK, 2018, pp. 12, 26, 32).

The COVID-19 crisis vs. the state budgetary bank tax revenues in 2020
The year 2020 brought a radical change in the conditions determining the accumulation of state budget revenue. The key factor here was the macroeconomic situation caused by the COVID-19 pandemic. Due to the incidence of the coronavirus, a number of restrictions were introduced regarding the functioning of the Polish economy. The pandemic has changed the behaviour of both consumers and enterprises, which had a negative impact on the dynamics of economic growth. In 2020, nominal GDP increased by only 1.3%, whereas in 2019 it was still at 8.1%. In 2020, on the other hand, real GDP ultimately decreased by 2.7%. As a consequence, the state budget revenues that year were below the amounts planned in the implementation schedule of the Budget Act. At the end of October 2020 the act was amended, reducing the total income forecast by 8.42% (Rada Ministrów, 2021, pp. 61-63).
The revenues from the tax on certain financial institutions were revised downwards as well -from PLN 4 878 million to PLN 4 740 million, a reduction by 2.83% (Rada Ministrów, 2021, p. 63). Moreover, the analyses presented by the National Bank of Poland (NBP, 2020b) indicate that the magnitude of the shock caused in the financial sector by the COVID-19 pandemic was significant, while the level of uncertainty as to its consequences was high. A decrease in the demand for bank loans was also pointed out. This phenomenon resulted, among others, from the fact that in Poland, due to the restrictions introduced, enterprises receive advantageous government support (Rada Ministrów, 2021, p. 80).
Despite the above-mentioned difficulties, the bank tax revenues, as indicated in Figure 3, increased nominally in 2020 compared to the previous year. Thus the main research hypothesis has been confirmed. The bank tax revenues amounted to PLN 4 822 million and were higher, compared to the forecast included in the amendment to the Budget Act, by approximately PLN 82 million, i.e. 1.7%. Compared to 2019, the income from the bank tax was higher by approximately PLN 122 million, i.e. 2.6% in nominal terms. In real terms, however, it decreased by only 0.8% (see Figure 3). The nominal increase resulted from the fact that the factors increasing the tax base prevailed, despite the increase in the value of the equity and the value of Treasury securities lowering this base (Rada Ministrów, 2021, p. 80). At the end of 2020, the banks' exposure to Treasury securities and Treasury guaranteed bonds amounted to PLN 464 5 billion and accounted for over 22.5% of the sector's asset value. At the end of 2019, the share of this group of assets was still at the level of 18.5%. Over the year it increased by nearly 4 pp (NBP, 2021a, pp. 7, 73  It is worth mentioning that at the end of 2020 the banking sector's assets were 17.8% higher than the year before. The increase in the balance sheet total resulted mainly from the high dynamics of the bank customers' deposits, especially those made by enterprises. A large amount of the funds received by these entities as part of successive Financial Shields, namely as government aid during the lockdown period, was kept in bank accounts. The reason for this was the unstable macroeconomic situation and the repayable nature of part of the aid provided under the shields (Raport o sytuacji…, 2021, pp. 124-125). The higher liquidity of enterprises also resulted from the fact that, due to the lockdown, these entities limited both investments and economic activity. In June 2020, the amount of non-financial enterprises' deposits in banks increased by as much as 30% year on year (Rudke, 2020), while at the end of 2020 the growth dynamics of these deposits amounted to 19.8% (Raport o sytuacji…, 2021, p. 195). Banks invested these funds in Treasury securities which are exempt from the bank tax, or in the Polish Development Fund (Polski Fundusz Rozwoju, PFR) and the Bank Gospodarstwa Krajowego bonds. Although the bonds issued by these entities are not excluded from the bank levy tax base, banks received higher interest on them (Rudke, 2020).
In 2020, owing to, among others, the increase in the banks' exposure to Treasury bonds, the sovereign-bank nexus increased (NBP, 2020b(NBP, , pp. 7, 73, 2021a. As already emphasized, one of the three components making up this index entails the sensitivity of banks' equity to the credit risk associated with Treasury bonds. In 2020, its value increased by 0.10 points, i.e. up to 0.28, whereas the entire index reached its highest value at 0.90 (see NBP, 2021b). Thus this index, as a whole, was heading towards its maximum (for more, see: NBP, 2021a, p. 73).
Furthermore, in 2020 the value of Treasury securities and Treasury guaranteed bonds exceeded the banks' equity two and a half times, which increased these institutions' sensitivity to changes in bond prices. The factors that will limit this risk, however, are the relatively good prospects for the fiscal situation in Poland, as well as the bond purchase programme launched by the NBP in March 2020 (for more, see : NBP, 2020b: NBP, , pp. 57-58, 2021a. This programme gains particular importance when the banks' other burdens, especially those associated with credit risk, will possibly accumulate (NBP, 2020b, p. 7).
The economic slowdown and the anti-crisis measures undertaken by the Polish government with the aim of mitigating the effects of the pandemic have led to a deterioration of Poland's public finances 10 in 2020. The scale of the anti-crisis measures were estimated at 4% of GDP that year, which contributed to an increase in the public finance sector's expenditure. At the same time, the economic downturn weakened the growth of the sector's income. This resulted in an increase in the public finance sector's deficit indicator calculated using the ESA 2010 methodology, from 0.7% of GDP in 2019 to 7% of GDP in 2020. Poland's public debt, as per the ESA, increased at that time by as much as 11.9% of GDP, reaching 57.5% of GDP (NBP, 2021a, pp. 13-14).
The increase in the public finance sector's deficit resulting from the measures limiting the direct economic impact of the pandemic, as well as the reduction of the banking institutions' lending activity, will increase the share of Treasury securities in these institutions' portfolios. The strength of the links between the public and the banking sectors (sovereign-bank nexus) will therefore increase both ways: on the banking sector side -through the increased involvement in the Treasury securities that are exempt from the bank tax and thus may crowd out bank loans during an economic downturn; and on the government side -through the increased importance of the credit risk in Treasury security valuation. NBP analyses, however, show that in 2020 the government's credit risk valuation increased to a much lesser extent, compared to the period of the 2008 global financial crisis (NBP, 2020a, pp. 72-73).
Under traditional circumstances, the aforementioned increase in Poland's public debt could adversely affect the valuation of Treasury securities. This indicator, nevertheless, is still relatively low compared to many European Union countries 11 . Moreover, the phenomenon of public debt increase, mainly caused by the response to the COVID-19 pandemic, is currently quite common worldwide. Therefore, individual EU countries' ratings and the market valuation of their credit risk do not signal a deterioration of the Polish State's creditworthiness due to the current or expected increase in public debt (for more, see: NBP, 2020b, pp. 56-57).
As the International Monetary Fund's consultation report on Poland indicates (IMF, 2021), maintenance of the private sector's access to credit will be of extreme importance for the recovery of the Polish economy after the COVID-19 pandemic. The authors of the report indicated, among others, the need for changes in the bank tax policy in order to support lending during the phase of economic recovery. For this tax to be less distortive to the banks' activity, they suggested the introduction of a quasi-Value Added Tax on the profits and remuneration in these institutions (IMF, 2021, p. 23). The Polish Bank Association, in turn, appealed for the exemption of, among others, newly granted investment loans and the loans earmarked for investments under the "European Green Deal" programme from the bank tax (for more, see: ZBP, 2021). 11 According to the 30 th of April 2021 forecast, the average public debt indicator for the entire EU (27 countries) was 92.4% of GDP at the end of 2020, as per ESA 2010. In Poland this indicator was set at 57.5% of GDP. The EU countries' debt ranking classified Poland in 12 th place, counting from the lowest level of the indicator. Estonia had the lowest public debt indicator (18.2% of GDP), while Greece had the highest (205.6% PKB). A level of public debt above 100% of GDP was observed in other six member states (for more, see: EC, 2021a, p. 169).

Conclusion
Most of the European Union countries introduced the bank tax as compensation for the public aid granted for the restructuring of financial institutions during the financial crisis. Poland adopted this tax with a considerable delay, for a purely fiscal purposeit was imposed on the assets of certain financial institutions and is paid by the largest banks, part of the insurance sector, and some loan companies. These entities pay the bank levy every month, while the tax base is the surplus of the total value of these entities' assets. Banks can lower this base by, among others, the value of the Treasury securities purchased. The legal structure of the bank tax also includes exemption thresholds, which means that neither credit unions nor cooperative banks pay the tax.
In the first four years of the bank tax in Poland, the proceeds from this levy totalled over PLN 17 billion, which accounted for slightly more than 1% of the state budget revenues. The level of the inflows was stable, with a slight upward trend. The grounds for this trend lie in the increase in the assets of the banking sector. The effectiveness of bank tax revenue planning varied. In 2016-2017, significant deviations from the plan occurred, nevertheless these often accompany the introduction of sectoral taxes. In the following years, i.e. 2018-2019, these variations were much smaller. The banking sector had a dominant share in the proceeds from the bank levy of slightly over 85% on average. Domestic insurance and loan companies paid a total of less than 15%. Another bank tax after-effect for the state budget was the increase in the banks' demand for Treasury securities. By investing in Treasury securities, banks reduced the tax base and, in consequence, the tax itself. This was also beneficial for the government sector, which could finance the budget expenses from the sale of these securities. In 2016-2019, the value of these securities in the banks' portfolio increased by a total of 72.4%, reaching PLN 333.2 billion at the end of this period. The Supreme Audit Office, however, did not find the Polish banks' use of Treasury securities as tax optimization. Nevertheless, the above-mentioned relief deepened the complexity of the relationship between the government and the banking sectors. The strength of this complexity is measured by the sovereign-bank nexus index. In 2016, the year of the introduction of the bank levy, a significant increase occurred in the variable making up this index, i.e., in the variable expressing the sensitivity of banks' equity to the credit risk of Treasury bonds. What is more, in the analysed period, the Ministry of Finance undertook steps aimed at preventing bank tax evasion. As a result, the taxpayers submitted corrections to their tax returns. Certain situations emerged when individual interpretations were changed in the taxpayers' favour after the court judgments in this respect became final. The overall assessment resulting from the audit carried out by the Supreme Audit Office with regard to the Minister of Finance's supervision over the collection of bank tax in the initial period of its operation was positive.
The year 2020 brought a significant change in the conditions of state budget revenue collection, including bank levy revenues. The outbreak of the COVID-19 pandemic resulted in the introduction of a number of restrictions in the functioning of the economy. This had a negative impact on the dynamics of the economic growth. As a consequence, the budget revenues were below the amounts planned in the schedule of the Budget Act, which was therefore amended in October 2020, for example the bank tax revenue forecast was lowered. Despite the above difficulties, in 2020 these revenues increased by 2.6% in nominal terms, compared to 2019. Thus the main research hypothesis was confirmed. The bank tax revenues amounted to PLN 4.8 billion and were higher compared to the forecast included in the amendment to the Act, by 1.7%. This increase resulted from the predominance of the factors increasing the tax base. One of such factors was primarily the high dynamics of the deposits made by bank customers, especially enterprises. A large share of the funds received by these entities as part of the successive financial shields introduced as government aid during the freezing of the economy, was kept in bank accounts. Banks invested these funds, among others, in Treasury securities, which are excluded from the bank tax. This caused a further increase in the sovereign-bank nexus index, which reached its highest value at the time. The analyses carried out by the National Bank of Poland, however, showed that in 2020 the Polish government's credit risk valuation increased to a much lesser extent compared to the global financial crisis in 2008. In order to support lending during the economic recovery phase, after the COVID-19 crisis, changes to the bank tax rules are required. For this levy to be less disruptive to the banks' activity, it is necessary to consider tax base changes, from assets to a quasi-Value Added Tax on the profits and remuneration in these institutions. The exemption from the bank tax on newly granted investment loans or the loans intended for investments under the "European Green Deal" programme should also be considered.