The recent unexpected collapse of the Silicon Valley Bank in the US calls for a better understanding of the sources of bank runs. In a recent working paper, my co-authors (Raja Reddy Bujunoori and Nishant Kashyap) and I show that a sudden increase in bank transparency can trigger runs.

Our main finding is that the recent asset quality review of banks caused a run by mutual funds on NBFCs having exposure to borrowers (from banks) with significant adverse audit findings.

Our papers address an old debate among scholars about whether banks should be fully transparent or whether a level of opacity in bank assets is desirable.

The benefits of transparency are well known: it enables investors and depositors to take informed decisions and regulators to act in a timely manner.

Since bank managers cannot hide bad assets under a transparent regime, the risk of a sudden collapse of banks due to bad asset quality reduces.

A 2016 Journal of Accounting Research paper by Viral Acharya and SG Ryan highlights the importance of bank transparency in ensuring financial system stability.

Value of deposits

What, then, is the argument in favour of optimality of some level of bank opacity? Tri Vi Dang, Gary Gorton, Bengt Holmström, and Guillermo Ordonez highlight the benefits of some level of bank opacity in their 2019 American Economic Review paper. To appreciate their argument, it is important first to understand that bank liabilities (deposits) are money like instruments.

In other words, a depositor treats his/her bank deposits on par with money and behaves as if it does not fluctuate in value regularly.

Most transactions happen through a transfer of bank deposits between different depositor accounts. When a customer pays a restaurant bill using UPI, he/she is transferring deposits from his/her account to the account of the restaurant owner in the same or other banks.

The fact that both the depositor and the restaurant owner treat the deposits at par value enables this transaction. Imagine using equity shares or some listed bonds for transactions. It is hard to arrive at the required number or value of shares or bonds in exchange for a meal as the value of these instruments keeps fluctuating even in the short run. Entering into medium- or long-term contracts is even harder.

Regarding bank deposits, everyone normally behaves as if they do not fluctuate in value. In spirit, this is not true. The assets which the bank deposits finance do fluctuate in value, and therefore, in theory, even the value of deposits also should fluctuate.

However, since bank assets are generally opaque, banks do not reveal all the information they have about their loans, depositors behave as if bank deposits do not fluctuate in value. The availability of deposit insurance and implicit protection by governments aid this phenomenon further.

Thus, opacity allows bank deposits to be used as money and enables money creation by banks. In a highly transparent banking regime, along with bank assets, bank deposits will also fluctuate in value and lose their place as money-like instruments.

Asset quality review

Here is where the role played by AQR becomes important: it exposed the quality of bank assets. During the course of AQR, the declared NPAs skyrocketed from about 3 per cent to nearly 11 per cent of bank assets.

The NPA levels for some banks crossed the 20 per cent level. This meant that banks reached negative capital levels. In other words, the true value of bank deposits was below par.

More importantly, the fact that the deposits were less than par was known to all, including the depositors. This created a good setting for triggering a run.

However, when it comes to bank deposits, there are economic forces that could counterbalance the incentives to run. First, most banks in India are government owned. The chances of government banks eventually running out of cash, irrespective of the level of negative shocks to assets, are low. Second, in the past, the government and the regulators have not allowed even large private banks to fail — YES Bank being a case in point.

Finally, a large section of depositors, especially retail depositors, are passive and are not fully informed about events such as the AQR.

However, it is crucial to note that the borrowers who got exposed as poor-quality borrowers in the AQR also borrow from non-banking finance companies (NBFCs). In other words, there could be apprehension that deposits and investments with NBFCs are also below par.

The NBFC factor

Our working paper referred to above argues that none of the three conditions described above applies to NBFCs: most NBFCs are not government-owned, they are not generally bailed out when in trouble, and the depositors are generally well-informed.

Our paper shows that debt mutual funds that had investments in NBFCs which, in turn, had exposure to borrowers revealed to be poor in the AQR, witnessed significant redemptions immediately after the AQR. The mutual funds, in turn, withdrew from their liquid investments with the NBFCs, creating a run-like situation for NBFCs. In the absence of a bailout, these affected NBFCs failed.

Eventually, the RBI and the government had to bail out the NBFCs through special liquidity facilities offered through banks. Thus, our paper shows that the NBFC crisis that the country witnessed was possibly an unintended consequence of increased bank transparency.

Our findings show that it is important that bank regulators and policymakers consider the possibility of a run somewhere in the financial system as a potential unintended consequence of their proposed actions to increase bank transparency.

The writer is an Associate Professor with Indian School of Business