 "How and how not to analyse bank financial performance")
Come quarterly/annual bank financial results, investors and readers of financial newspapers are all agog over some analyst gushing ‘Bank A’s net profit rises 35%’ or some other analyst emoting ‘Bank B’s NIM (net interest margin) is highest at 6%’. All these reports are taken by readers and investors as holy grail suggesting that the banks concerned have been exceptionally efficient and profitable. This need, and may, not at all be so. Before seeing why, it would only be instructive to consider the business model of a typical competitive, efficient, safe and sound bank.
A bank is typically characterised by relatively high financial leverage, which, in turn, is measured by what is known as the equity multiplier, which, in turn, is nothing but the bank’s total assets divided by equity/shareholder funds. Multiplying this leverage (equity multiplier) by Return on Assets (RoA) gives Return on Equity (RoE) for a bank. Historically, competitive, efficient, safe and sound banks have had an average RoA of about 1% and a reasonably safe equity multiplier of about 15, implying an average market-competitive equilibrium RoE of about 15%.
In the recent period, the Indian banking system has had a leverage of about 13 to 14 times. Significantly, and hearteningly, to the credit of the Reserve Bank of India and the Indian banking sector, this corresponds to an average leverage ratio (inverse of equity multiplier) of 7% plus, which is way higher than the 3% mandated by the new Basel-III capital rules to be complied with in 2018. In other words, the Indian banking sector is already more than 2.5 times compliant on this critical Basel III parameter.
As regards the myth of NIM being a key measure of profitability, NIM by, and in, itself conveys nothing more than what it apparently does - it is just the difference between interest earned and interest expended, as a percentage of a bank’s assets. It is just a means to an end and not an end in itself. For NIM to make any sense, it needs to be analysed further beyond what it is by considering ‘NIM-RoA’. For example, if NIM is 6% and RoA is zero, then the RoE will also be zero. Thus, it is obvious to see that this nominally very high NIM only establishes that the bank is neither competitive, efficient, safe nor sound. Even if RoA was 2%, then ‘NIM-RoA’ will be 4% (6%-2%). And this bank will be far less efficient and competitive than a bank whose NIM is 3%, and RoA is 1%, and, therefore, ‘NIM-RoA’ works out to 2%. This is because non-interest cost of intermediation of the higher NIM bank is twice that of the lower NIM bank and it is precisely this twice as large ‘NIM-RoA’ and its reasons through its granular analysis and dissection that should engage the attention of bank analysts and investors.
For it is this ‘NIM-RoA’ that subsumes all non-interest expenses such as taxes, salaries, operational expenses, loan loss provisions, marked-to-market provisions, write-offs etc. And this ‘NIM-RoA’ becomes even more significant, if the reported gross NPAs (non performing assets) are unusually low. Therefore, in the above example, the bank with lower NIM-RoA will be twice as efficient and competitive as the one with higher NIM-RoA because the former maximises value for all stakeholders viz depositors by way of higher deposit interest rates, borrowers by way of lower borrowing costs, shareholders by way of given RoA and market-competitive equilibrium RoE.
And significantly, there is no effect on the rigour, applicability and generality of the above analytical framework even if non-interest income is not insignificant for all that need be done is just add this income, as a percentage of assets , to NIM- RoA, and only refer to this, if you will, as Net Income Margin rather than Net Interest Margin. This would, for a given RoA and RoE, only make NIM even wider and, as stated before, in combination with the unusually low gross NPAs for such banks, would make NIM-RoA look, if anything, even more questionable in the sense that what is absent in the unusually low Gross NPAs is very likely present in the unusually wide NIM-RoA.
Finally , coming to too much being made of, say 25% to 35% growth in net profits, this too needs to be regarded with circumspection for these numbers need to be adjusted for the growth in balance sheet/assets and not just considered in isolation and on a standalone basis. For if net profit grows at 35%, on a yoy ,or a CAGR, basis and assets/ balance sheet also grow by say 35%, then there is really nothing to write home about for the bank in question as it has been no more, and no less, efficient and profitable than before.
Another equally insightful way to see this is in terms of change in RoA. For example, if previous RoA is 1%, then there is no change in RoA as the RoA also remains unchanged at 1% for 35% growth both in net profits and assets/balance sheet. On the other hand, if for a 35% growth in net profit, assets/balance sheet grow by, say 25%, then the bank has been more efficient and profitable only to the extent of (1.35/1.25-1)*100 i.e.+ 8%, and not 35%, as bank analysts would unwittingly have readers and investors believe. In this case, RoA increases fom 1% to 1.08% (1*1.08) only. Also, significantly, and equally, if assets/ balance sheet grow by 40%, then the so-called nominal profit growth of 35% will translate into a less efficient and less profitable performance of (1.35/1.40-1)*100 i.e - 3.5% and not 35% as RoA will decline to 0.96% from 1% previously although absolute net profit increased by 35%. This then is the conceptually robust and technically rigorous nuts-and-bolts way of how bank analysts and investors must dissect bank financial performance and judge true and fair value of banking stocks for value investing/buying.
The author is former executive director, Reserve Bank of India
A bank is typically characterised by relatively high financial leverage, which, in turn, is measured by what is known as the equity multiplier, which, in turn, is nothing but the bank’s total assets divided by equity/shareholder funds. Multiplying this leverage (equity multiplier) by Return on Assets (RoA) gives Return on Equity (RoE) for a bank. Historically, competitive, efficient, safe and sound banks have had an average RoA of about 1% and a reasonably safe equity multiplier of about 15, implying an average market-competitive equilibrium RoE of about 15%.
In the recent period, the Indian banking system has had a leverage of about 13 to 14 times. Significantly, and hearteningly, to the credit of the Reserve Bank of India and the Indian banking sector, this corresponds to an average leverage ratio (inverse of equity multiplier) of 7% plus, which is way higher than the 3% mandated by the new Basel-III capital rules to be complied with in 2018. In other words, the Indian banking sector is already more than 2.5 times compliant on this critical Basel III parameter.
Also Read
In this context, another key parameter – the net interest margin (NIM) -- which is the difference between interest earned and interest expended as a percentage of a bank’s assets is relevant. Collectively for Indian banks, NIM has varied between 2.5% to 3% in recent times. If we deduct RoA from NIM, we get what can be called the Non-interest Cost of Intermediation. In fact, this ‘NIM-RoA’ (NIM minus RoA) is this critical parameter which, for a given RoA derived, in turn, from a given RoE and equity multiplier, must be the dharma/mantra of a role model bank management to minimise for maximising returns to depositors and/or minimising costs to borrowers. Thus, either way, minimisation of NIM-RoA delivers value to all stakeholders viz, shareholders, insured and uninsured depositors, borrowers, taxpayers, in particular, and the real economy, in general.
As regards the myth of NIM being a key measure of profitability, NIM by, and in, itself conveys nothing more than what it apparently does - it is just the difference between interest earned and interest expended, as a percentage of a bank’s assets. It is just a means to an end and not an end in itself. For NIM to make any sense, it needs to be analysed further beyond what it is by considering ‘NIM-RoA’. For example, if NIM is 6% and RoA is zero, then the RoE will also be zero. Thus, it is obvious to see that this nominally very high NIM only establishes that the bank is neither competitive, efficient, safe nor sound. Even if RoA was 2%, then ‘NIM-RoA’ will be 4% (6%-2%). And this bank will be far less efficient and competitive than a bank whose NIM is 3%, and RoA is 1%, and, therefore, ‘NIM-RoA’ works out to 2%. This is because non-interest cost of intermediation of the higher NIM bank is twice that of the lower NIM bank and it is precisely this twice as large ‘NIM-RoA’ and its reasons through its granular analysis and dissection that should engage the attention of bank analysts and investors.
For it is this ‘NIM-RoA’ that subsumes all non-interest expenses such as taxes, salaries, operational expenses, loan loss provisions, marked-to-market provisions, write-offs etc. And this ‘NIM-RoA’ becomes even more significant, if the reported gross NPAs (non performing assets) are unusually low. Therefore, in the above example, the bank with lower NIM-RoA will be twice as efficient and competitive as the one with higher NIM-RoA because the former maximises value for all stakeholders viz depositors by way of higher deposit interest rates, borrowers by way of lower borrowing costs, shareholders by way of given RoA and market-competitive equilibrium RoE.
And significantly, there is no effect on the rigour, applicability and generality of the above analytical framework even if non-interest income is not insignificant for all that need be done is just add this income, as a percentage of assets , to NIM- RoA, and only refer to this, if you will, as Net Income Margin rather than Net Interest Margin. This would, for a given RoA and RoE, only make NIM even wider and, as stated before, in combination with the unusually low gross NPAs for such banks, would make NIM-RoA look, if anything, even more questionable in the sense that what is absent in the unusually low Gross NPAs is very likely present in the unusually wide NIM-RoA.
Finally , coming to too much being made of, say 25% to 35% growth in net profits, this too needs to be regarded with circumspection for these numbers need to be adjusted for the growth in balance sheet/assets and not just considered in isolation and on a standalone basis. For if net profit grows at 35%, on a yoy ,or a CAGR, basis and assets/ balance sheet also grow by say 35%, then there is really nothing to write home about for the bank in question as it has been no more, and no less, efficient and profitable than before.
Another equally insightful way to see this is in terms of change in RoA. For example, if previous RoA is 1%, then there is no change in RoA as the RoA also remains unchanged at 1% for 35% growth both in net profits and assets/balance sheet. On the other hand, if for a 35% growth in net profit, assets/balance sheet grow by, say 25%, then the bank has been more efficient and profitable only to the extent of (1.35/1.25-1)*100 i.e.+ 8%, and not 35%, as bank analysts would unwittingly have readers and investors believe. In this case, RoA increases fom 1% to 1.08% (1*1.08) only. Also, significantly, and equally, if assets/ balance sheet grow by 40%, then the so-called nominal profit growth of 35% will translate into a less efficient and less profitable performance of (1.35/1.40-1)*100 i.e - 3.5% and not 35% as RoA will decline to 0.96% from 1% previously although absolute net profit increased by 35%. This then is the conceptually robust and technically rigorous nuts-and-bolts way of how bank analysts and investors must dissect bank financial performance and judge true and fair value of banking stocks for value investing/buying.
The author is former executive director, Reserve Bank of India