پارادوکس در بیمه سپرده بیمه تحت عملکرد آماکوداری در سیستم بانکداری ژاپن

Changing deposit insurance from full to limited in 2005 is expected to discipline Japanese banks’ behaviour because depositors will start monitoring their banks. However, this discipline effect may be overturned due to the amakudari practice in the Japanese banking system where regulatory officials obtain post-retirement jobs in private banks. We consider a signalling game where depositors and banks have asymmetric information regarding banks’ riskiness, and banks use amakudari officials to signal their riskiness. In order to create more post-retirement employment opportunities, the regulatory authority may weaken prudential regulation, which results in less discipline in the banking industry.

The Japanese economy has witnessed a large-scale regime change in the regulatory framework of the financial industry since the financial crisis that arose in 1997 with the collapse of large financial institutions. For example, the once-mighty Ministry of Finance (the MOF) was deprived of the regulatory authority over the financial industry. The authority is now in the hand of the newly established regulatory organisation called the Financial Services Agency (the FSA). Unsurprisingly, the deposit insurance system was also forced to change because the old system was built on the assumption of the “convoy system.” The most important feature of the new deposit insurance system is an introduction of limited deposit insurance or the so-called “payoff system.” The Japanese deposit insurance system can be split into three phases: implicit full deposit insurance, explicit full deposit insurance, and limited deposit insurance.1 The limited insurance system nominally existed as early as in 1971 and covered deposits up to one million yen (1st phase). This figure was raised to ten million yen in 1986. However, under the “convoy system,” all types of deposits were effectively fully insured. That is, whenever banks became insolvent, the MOF rescued them by arranging a takeover by healthier banks. In essence, it was an implicit full deposit insurance system, which was based on the assumption that the failure of banks was sporadic and small in scale. However, following the spread of credit uncertainty after the collapse of credit co-operatives in December 1994 and other incidents, the MOF decided to freeze the limited insurance system (although it was nominal and payoff had never been exercised) until April 2001, and the implicit full insurance system turned into the explicit (2nd phase). However, the freeze eventually was partially lifted with the cap being imposed upon saving-type deposits, such as time deposits, in April 2002, while ordinary deposits were reprieved for two more years until March 2005. From April 2005, the limited deposit insurance system has been in place (3rd phase). The limited deposit insurance system is expected to serve for two well-known objectives. First, depositors play an active role in disciplining banks’ behaviour in line with depositors’ interests. That is, limited deposit insurance gives depositors strong incentives to monitor banks. For depositors with large-scale deposits, lenient bank monitoring may end up with losing most of their deposits. Second, the limited cover of deposits prevents depositors from free-riding behaviour. Under full insurance, people are willing to deposit their funds in banks that offer higher interest rates regardless of their riskiness. In fact, behaviour of this kind was observed in the Cosmo Credit Union case in the 1990s in Japan. Limited deposit insurance surely discourages depositors’ free-riding behaviour. Limited deposit insurance also has tremendous effect upon banks themselves. Banks now have to take into account depositors’ response to their actions. Under the full deposit insurance system, there was no reason for depositors to be concerned with riskiness of banks because their deposits were fully insured in any event. For banks, this meant that they could take any risks because it would not have changed depositors’ behaviour. However, under the new deposit insurance system, banks’ behaviour directly affects the interests of depositors through the partial insurance. Banks that take excessive risk may face a massive outflow of deposits and may be forced to close down. Hence, the limited deposit insurance system makes depositors an important player in the financial sector in Japan. As a consequence, depositors should become extremely concerned about behaviour and quality of banks. Depositors will try obtaining as many pieces of information as they can regarding banks’ riskiness, profitability and financial health, which potentially influence their deposits. Amongst those, our focus is the relationship between banks and the regulatory authority. Namely, we focus upon a well-known custom in the Japanese banking industry — the amakudari practice. For a long time, many Japanese regional banks have been accepting retiring officials from the financial regulatory authority as senior managers, members of the managerial board or presidents. This practice is often referred to as the amakudari practice and those officials are called amakudari officials. The objective of this paper is to examine this amakudari practice and predict its possible effects upon the Japanese banking system under the new deposit insurance system. Our innovation in this paper is that we let depositors extract information on banks’ quality by observing banks’ attitude in hiring amakudari officials. Again, we stress that this “bank-monitoring role” played by depositors will become important under the new deposit insurance system in Japan. Knowing depositors are monitoring, in turn, banks use amakudari officials as a signalling device to manipulate depositors’ assessment about their quality. That is, banks with different degrees of risk can use amakudari officials as a signal of their riskiness. In essence, our paper puts emphasis upon strategic interactions between banks and depositors through the amakudari practice, provided an active role played by depositors under the new deposit insurance system. Naturally, our paper is the first attempt to analyse the amakudari practice within a game theoretic framework (because under the previous systems, depositors’ response was not of banks’ concern), where all the economic literature regarding the amakudari practice has been purely empirical ( Horiuchi & Shimizu, 2001; Suzuki, 2001, van Rixtel, 2002 and Yamori, 1998). We consider a game in which there exists asymmetric information between depositors and banks where the riskiness of each bank is private information. That is, riskiness is observable for individual banks but not observable for depositors. Our model predicts various equilibria under the new deposit insurance system in Japan, including ones where amakudari officials are hired even when they are unproductive for banks. We will also see equilibria under which the amakudari practice is no longer observed. It will be shown that the type of equilibria depends upon (1) productivity of officials; (2) depositor's risk aversion; and (3) the riskiness of banks. We will also show that, in contrast to what some of the existing empirical literature has found: (1) the existence of the amakudari practice may give incentives for the regulatory authority to weaken prudential regulation in the banking industry; and (2) the amakudari practice as an incentive scheme for the officials in the regulatory authority can be important. 2 Our first finding is especially interesting in the sense that whilst limited deposit insurance is expected to discipline the banking industry, paradoxically with the co-existence of the amakudari practice in the Japanese banking industry, it may not work in the way as one might expect. The remainder of the paper is structured as follows. Section 2 overviews the amakudari practice, discusses its functions, and reviews the previous literature. In Section 3, we develop our model and discuss the paradoxical effect of limited deposit insurance. Section 4 concludes.

We have explored the amakudari practice in the Japanese banking system, focusing particularly upon the relationship between banks and depositors, which appears to become extremely important under the limited deposit insurance system, which has been introduced recently. Our signalling model predicts various equilibria depending upon (1) productivity of officials; (2) depositor's risk aversion; and (3) the riskiness of banks, and we have related these equilibria to the existing literature. We have explained a theoretical possibility that the amakudari practice may weaken the regulatory authority's general prudential regulation in the banking sector under the limited deposit insurance system. Our finding is similar to but is different from what the existing empirical literature found. Whilst the literature only put emphasis upon undesirable effects of the amakudari practice upon individual banks, which accepted the officials, the potential harm suggested by our analysis is more serious and widespread across the whole banking industry. Our game-theoretic model found a route for the amakudari practice to have an effect upon prudential regulation, which has never been investigated in the existing literature. More importantly, our research is the first attempt to explore the implication of the amakudari practice under the limited deposit insurance system in Japan where depositors are expected to share the role of bank monitoring with the regulatory authority. The novelty of the implication of our model is the paradox of the limited insurance system under the amakudari practice. It is obvious that the objective of the introduction of the limited insurance system is to give depositors some incentive to discipline banks. The limited deposit insurance system is expected to strengthen the effectiveness of disciplining banks together with the regulatory authority's prudential regulation. However, our analysis shows the case that the limited insurance system undermines the incentive for the regulatory authority in bank monitoring in presence of the amakudari practice. Interestingly, the driving force of this uncomfortable result is the information activity that depositors are expected to exercise under the limited deposit insurance system. Depositors try to collect and use all available information to judge whether their banks are safe or risky. Our analysis shows that there may be opportunities for the regulatory authority to take advantage in the signalling game between banks and depositors. Our model predicts that the combination of the limited deposit insurance system and the amakudari practice may result in the trade-off between depositor and the regulatory authority's bank disciplining efforts. Under the limited insurance system, depositors are expected to play a complementary role in bank monitoring. Ironically, this may not be the case in Japan where limited deposit insurance system and the amakudari practice will coexist, because the regulatory authority may attempt utilising depositors’ monitoring activity to increase its employment opportunities in the banking sector. We suggest two possible solutions to the problem. As we have examined, unproductive officials are the major source of social suboptimal outcomes. Therefore, if we know that officials are unproductive, a prohibition of amakudari appears to be the most straightforward solution. The other solution is to amend the way officials are trained during their careers so that they become productive in private banks. They had better be trained more as banking financial market “specialists” with expertise in finance and practical knowledge of the financial markets, than being trained as “generalists” who are only required to have the ability to deal with general administration. The harmful effect of the amakudari practice we have analysed is purely theoretical at this stage. Although it is unlikely that the full effect of the limited deposit insurance system discussed in this paper will be immediately apparent, empirical studies should be conducted to verify our prediction. This work is left for the interested reader.