A complete solution to India’s bad debt problem should address recovery from defaulted accounts and optimization of capital requirement for banks without aggravating the government’s fiscal position. Additionally, the solution should be socially and politically acceptable. Given the size of the problem and the requirements for the potential solutions, one needs to look at a broad set of solutions and not exclude any before full evaluation. One much-proposed, much-maligned solution is that of the “bad bank". It is much like chemotherapy: no one likes it, but in difficult times, it has to be recommended despite the side effects.

The idea of a bad bank faces such widespread resistance that detailed deliberation on a well-structured bad bank rarely happens. Therefore, when a bad bank-like solution is actually proposed, it is often suboptimally structured. The narratives used to discredit the bad bank idea revolve around how it will entail the government using taxpayers’ money to bail out banks—the same banks that apparently risked their depositors’ money while making ill-conceived or ill-fated big-ticket loans.

The importance of “narrative" in policy and economic decision-making cannot be overstated. Narratives sometimes overtake economic rationale. George Akerlof and Dennis J. Snower, in a paper titled Bread And Bullets, focus on this element. Narratives tend to be simplified accounts of events. This facilitates easy and intuitive understanding of an otherwise complex process or event. However, that certain narratives become popular does not mean that they are providing accurate or comprehensive explanations.

The anti-bad bank narrative stems from two popular models used to describe banking. One is the “financial intermediation" model and another is the “fractional reserve banking" model. According to the former, banks bring savers’ money to borrowers—deposits lead to credit. The fractional reserve model, meanwhile, starts with deposits and continues to treat bank as intermediaries. Here, a bank retains a small fraction (say 5%) of the deposit (say 100) and lends 95. That 95 ends up as a deposit in another bank, which retains 5% of it as reserve and lends 95% of 95.

After 100 rounds of this process, the system would have disbursed 1,888 of credit from an original deposit of 100. Of course, because of the accumulation of successive deposits (which got created after the first lending of 95) on a static basis, one would see systemic deposit rising in tandem with credit, creating an illusion of full conformity with the intermediation model. However, in essence, in this model, 100 of initial deposit enabled the banking system to create credit of 1,788. The credit creates purchasing power—creates money in the economy.

The fractional reserve model highlights that it is not a bank alone but the system that creates money from the original deposit. However, it fails to explain the banking paradigm in countries such as the UK or Sweden where there are no deposit reserve requirements.

Of course, the oldest model to explain banking is the “credit creation" model. Economists such as John Maynard Keynes and Hyman Minsky, as well as several central bankers, have acknowledged this. After the 2008 crisis, the Bank of England virtually shouted from the rooftops, accepting the credit creation theory of banking. And the Swiss recently had a referendum on whether bankers should be allowed to create credit/money when they lend.

This third model argues that banks create deposits when they disburse credit. To explain, the accounting operations of “credit disbursement" are often reflected as liability/deposit account opened in the name of the borrower with the bank. The deposit adds to the money supply. According to the model, banks are not mere intermediaries. They create money when they decide to lend. In his paper A Lost century In Economics: Three Theories Of Banking And Conclusive Evidence, German economist Richard Werner presents empirical evidence rejecting the fractional reserve banking and financial intermediation theories.

Very few economic models can be rigorously validated with data and observations. Banking is possibly an admixture of all these three models, wherein banking systems with a sufficiency of capital and steady demand of credit are likely to see a dominance of the “credit creation theory". The other two models play an important but supporting role. Meanwhile, the dominant narratives driven by those two models obviate any solution that may be supported by the third model. That said, there is nothing to suggest that those two models are particularly successful in explaining banking. This makes the fact that they tend to block any solutions that may be supported by the credit creation model particularly detrimental.

As far as narratives go, according to the credit creation theory, banks, to an extent, are regulated franchises of the sovereign who create sovereign- backed money when they disburse credit. Credit crises thus call for explicit involvement of the government—else, the money creation function of the government gets affected. Money is an accounting measure to track who owns how much, with the claim backed by the sovereign. The appreciation of this narrative may lead to a well-structured bad bank, which will address the bad debt problem in totality. Of course, this doesn’t mean giving a free pass to borrowers and lenders who have misused the banking system.

Deep Narayan Mukherjee is a financial services professional and visiting faculty at IIM, Calcutta. Comments are welcome at theirview@livemint.com.