In the aftermath of the default of IL&FS in September 2018, the non banking finance companies (NBFCs) have been facing a significant amount of financial stress. Due to the increased cost of borrowing in the bond market, it has become difficult for the NBFCs to raise capital. In popular discourse, attention has been paid to the role played by the asset-liability mismatches in the balance sheets of the NBFCs in causing this crisis. In this article we highlight another problematic feature characterising the NBFC sector – potential underpricing of risk.
Background: A Decade Of Rapid Growth
The NBFCs have had a good run over the last few years. During the period 2009-2018, the aggregate loans extended by the NBFC sector grew at a CAGR of 17%. The sector registered the highest growth during the 2009-2014 period (24%). Several factors contributed to this good run. Given the burgeoning pile of non performing assets and capital shortfall to provide for the same, the public sector banks reduced their lending. Private sector banks' credit also slowed and remained concentrated in select segments. During 2009-2014, the CAGR of non-food credit extended by the banking sector was 12%. This created a credit supply shortage that presented the NBFCs with an expansion opportunity.
While banks reduced their exposure to the corporate sector, they continued to lend to the NBFCs. At the same time, increased investment of household financial savings in mutual funds meant that the NBFCs could raise funds from the debt mutual funds. The bond market emerged as the dominant source of financing for the NBFCs. During the period 2009-2018, the average share of bond market borrowing in the total borrowing of NBFCs was 51% whereas the share of bank borrowing was 26%.
Events in September 2018 effectively put a stop to the good run. Over the span of a few months, the growth of the NBFC sector rapidly declined and their margins came under pressure. This has resulted in some fundamental questions being raised about the NFBC business model. Here, we look at one crucial aspect of this business model.
Background: A Decade Of Rapid Growth
At the heart of any lending business is a rating arbitrage – the lender has to have, on average, better rating than the borrower so that its own borrowing cost is lower than what it charges the customer. The lender can also create a margin by running a maturity mismatch - borrowing short and lending long when the yield curve in upward sloping. Both these actions, credit arbitrage and maturity mismatch create risk for the NBFCs. In other words, the risk in the NBFC model has two components – market risk, conventionally called asset liability management (ALM) risk and credit risk that arises from lending to lower risk customers.
Following the IL&FS episode, there has been considerable discussion on the market risk embedded in NBFC balance sheets arising from the ALM mismatch problem. Regulatory actions have also focused on the market risk in terms of the reporting requirements for NBFCs and instructions to the rating agencies to consider market risk while rating the NBFCs. In our study, we take a closer look at the credit risk in the NBFC model.
NBFCs and Credit Risk
To understand the pricing of credit risk in an NBFC we have to look at two components – the pricing of credit risk when lenders such as banks and bond markets lend to an NBFC and the pricing of credit risk when the NBFC lends money to its customers. We call the former component ‘NBFC/ HFC premium’ while the latter ‘loan premium’. We define the NBFC premium as the difference between the average cost of borrowing for NBFCs and the 10-year government security yield. We define the loan premium as the difference between the average yield on loans given out by the NBFCs and the 10-year g-sec yield. The relative levels of these two premiums over time give a sense of the evolution of credit risk pricing in the overall NBFC sector.
For our analysis, we study the period 2009-2018. We focus on the post-global financial crisis period during which the NBFC sector grew steadily. We split our sample between NBFCs (non-deposit taking) and housing finance companies. We consider 11 largest NBFCs (excluding the government-owned ones) and 10 largest HFCs. We look at them separately because the nature of their businesses and hence the credit risk profiles are different. Together they represent about 80 percent of the entire NBFC sector.
We also look at the capital to loans ratio of the NBFCs/HFCs as a broad indicator of capital adequacy and the overall loan growth registered by the NBFCs/HFCs to get a sense of expansion of the sector. The chart below plot these variables for our sample of NBFCs/HFCs over the 10 year period.
Our observations from the charts can be summarised as follows:
- For our sample of HFCs, the HFC premium, the HFC loan premium and the HFC credit to loan ratio remained fairly stable over the period under study. The spread between the credit premium and loan premium also remained stable. This suggests that for the HFCs, the credit risk in general has been stable.
- For our sample of NBFCs, during the period 2009-2014, while the NBFC premium was stable, the loan premium steadily declined. The capital to loans ratio of the NBFCs remained stable throughout the period.
- During the period from 2009 to 2014, NBFC loans grew rapidly at a CAGR of 27%. The period average of loan growth for 2009 -2018 was 23%. Hence, during the time when the loan premium charged by the NBFCs declined, their loan books expanded rapidly.
This implies that while the lenders to the NBFCs kept their credit premium constant, the NBFCs themselves started reducing the premium they were charging their own customers. We do not find this trend for the HFCs.
The lowering of loan premium by the NBFCs could happen for two reasons. It could be that the NBFCs started lending to more creditworthy, lower risk customers. There is however no evidence to suggest the same. On the contrary anecdotal evidence suggests that in search of growth opportunities, NBFCs started lending to riskier customers. The alternative explanation for reducing loan premium is that NBFCs started systematically underpricing credit risk at a time of rapid growth. There was no commensurate improvement in the capital levels to support this underpricing.
The underpricing of credit may have been a result of intensifying competition in the sector which would drive down the lending rates or a correction of previous overpricing of credit or based on a perception that the overall risk in the economy was declining. Notwithstanding the reasons of underpricing, it would have made the NBFC business model riskier. Arguably the underpricing of credit helped the expansion of the NBFC sector during their high growth period.
Another way of interpreting the data is that despite the NBFCs relatively under-pricing credit risk on their loans and their capital levels not going up, the lenders did not penalise them by increasing the risk premium they charge the NBFCs. This could be because the bond market became a dominating source of financing for the NBFCs presumably due to increased inflows into debt funds and the bond market was relatively less discriminating in pricing credit risk.
If it is indeed the case that the NBFCs have been systematically underpricing credit risk as is apparent from our analysis, this could be a plausible reason behind the heightened risk perception about the sector in recent times. This would imply that ALM mismatches in the balance sheets are not the only problem in the NBFC business model which has some fundamental challenges that need to be addressed. Systematic mis-pricing of risk results in misallocation of capital. The shadow banking system in general will become safer and stronger when resources are efficiently allocated.
