Archive for the ‘Finance and Banking’ Category.


In the previous post, I covered several important variables in analyzing a bank. These factors are a good starting point in evaluating a financial institution, but they are not sufficient to arrive at a conclusion.

I am listing several additional criteria I consider personally, when analyzing these kinds of companies. Some of these factors are commonly used by other analysts, whereas some are of interest to me (even though others don’t care about them)

Growth – This is one of the top criteria used by a majority of the investors. A high growth trajectory (in deposit and advances) throws most analysts and investors into ecstasy. As some of you have realized, I like growth, but I am not a big fan. For most businesses, a moderate growth (between 12-15% per annum) is usually more sustainable, attracts lesser competition and provides good returns over the long term.

In the case of banks and other financial institutions, I am almost allergic to high growth. Financial institutions are highly leveraged institutions (read high debt) and as a result, a focus on growth can result in shaky loans which can haunt it in the future.

Take the example of ICICI bank – Don’t get me wrong on this one. I invested a miniscule amount in the bank IPO way back in the 90s and exited in the mid 2000s.I liked the bank service then (in late 90s the service was actually good!) and liked the way it was conducting its business.

However by mid 2000, the loan growth started increasing and my personal experience (and that of a few friends) of their underwriting standards (criteria to give you a housing or other loan) left me worried. They were much more lax in their standards than other banks. The bank has since then, slowed down its asset growth and is trying to work through its bad loans.

The key point of this story is this – An above average growth is good (though it does not guarantee conservative lending), but a high growth in a bank is a risky proposition. It may all work out in the long run, but I will not bet big on it.

Cost ratios

There are two costs ratios i look at closely when analyzing a bank or financial institution. The first one is borrowing costs, which I covered in the previous post. The other one is the operating cost ratio for the bank.

The operating cost ratio covers all the overheads of the bank such as salary for the employee, branch opening expenses, pension costs etc. I would prefer a downward trend in this number, unless the bank is expanding its network and is incurring the associated costs.

The new private banks such as Axis bank, which are expanding rapidly have an operating cost ratio in the range of 22-24%, where as the older private or public sector banks have this number in the range of 16-18%. I would expect the number to stabilize in this range for most banks as they expand their retail network and the growth slows down.

Credit deposit ratio

This is another important ratio to track. This is the ratio of deposits gathered by the bank to the amount lent out as loans. The RBI guideline is that this number should not exceed 75%. So if you see the number inching to 75%, the bank may have to resort to bulk deposits which are more costly than retail deposits – which means lower spreads and thus lower margins

In case you have a sneaky feeling that your bank is able to take a deposit at 7% from you and lend at 12% and make a nice spread on it – you are right. Banks have a nice thing going with its customers (you and me) – where they get money on the cheap and also charge money for all the other services they provide to us.

Yield on assets

One of the last commonly used ratios is the yield on assets – the return the bank makes on all the loans and other investments. I would like to see a high number, but too high a number could mean risky loans which could hurt the bank profits in the future.

So what is a high or low number? There are no absolutes here. The best option is to compare it across banks and get sense of this number. Currently, the average seems to be around 9.5-10%.

Let’s now look at additional factors which are not commonly followed

Contingent liabilities

I have yet to find a single report which talks of this. So what are contingent liabilities?

Think of these as possible costs, under certain circumstances (such as a particular level of interest rate changes) and hence they are called contingent. If you look at the balance sheet of a bank, all the open derivative and other contracts are included under this number.

For example, this number is around 3.2 Lac crore (yes not a typo) for axis bank which around 2 times their asset base. In a similar fashion this number has ranged between 3-4 for Yes bank and is as low as 25% for public sector and old private banks.

So whats the significance of this number? Does it mean a Yes bank or Axis bank is liable for 2 their total asset value (or 20 times networth ?).

The key point to remember is that these contingent liabilities are a notional value (total contract value) and not the amount which the bank would make or lose on these contracts. The amount which the bank can lose or gain is also provided in the notes to account.

If your head is hurting on hearing some these terms such as notional amount, derivative etc – I will not blame you. I cannot do justice to these topics in the post – you can easily Google it and find out.

The key point to remember is that contingent liabilities are off balance sheet risks (remember Lehman brothers and other investment banks ?). In good times, these derivatives help the bank in making money and are a nice source of ‘other income’ (the stuff which analysts like). However, if the market crashes or something nasty happens, then these contingent liabilities can kill the bank.

Does it mean Axis bank and Yes bank are risky banks ? Frankly I don’t know and an outside investor cannot evaluate the derivative book of a bank. However if you just use common sense in this case, a 25% ratio of contingent liability to asset (as in case of KV Bank) is definitely less risky than a 400% ratio in the case of Yes bank.

If you look at this ratio, the performance of several of the new gen, aggressive banks will make you pause and think

Other contingent liabilities

If you think, I have something against private banks, that is not the case. Public and old private banks have their cockroaches in their kitchen. These banks have pension and gratuity liabilities which have not been provided for. The RBI guideline requires the banks to provide these liabilities in phases and hence we are seeing the impact of these provisions on the results of the banks ( for ex: SBI in Q4).

I am however less worried about these kind of liabilities as they are not open ended and will be provisioned by the banks in the next 2-3 years.

No. of branches and ATM etc

I also like to track the growth in the number of branches, ATM and employees. The raw numbers alone are not enough. One also needs to look at the quality of the expansion – Is the bank expanding in clusters or is it making a thrust into the rural areas (which is good in the long term , though could hurt profits in the short term)

Technology adoption

There are no numbers for this factor. You have to read the annual report for the bank for the last few years and get a sense of how the bank is investing in the technology aspect of the business. Is the bank at the forefront of technology adoption or is it a few years behind the curve ?

Another easy way is to go to a local branch and see if you can get the various services such net banking, anywhere access etc from the bank.

Asset liability profile

Another data point which can be found in the notes to account. This table gives an indication, on whether the bank is exceedingly funded by short term deposits alone. It’s difficult for me to cover this topic in this post, but as a quick pointer – Higher the longer duration deposits, better the risk profile ( remember the term asset liability mismatch ? – if not, please look it up if you plan to invest in a bank)


We now come to a very important and the most difficult factor to evaluate. These are no numbers or tables to evaluate the bank’s management, but if you read the annual report and follow the management, you will get some sense of it.

For ex: Axis bank, ICICI and Yes bank have aggressive management which is looking at growing the bank on both the retail and lending side. HDFC has an aggressive management, but it is also very risk conscious. There are several old private sector banks, which have conservative managements which are growing the banks at a nice pace and with low risk.

Finally we have the public sector banks, where the management is essentially government deputed officers and so it’s difficult to get any picture as such banks (though in some cases there have been individuals who have done well, but then they are posted to some other institution)

Are you exhausted 🙂 ?

We have looked at all the factors which can be used to evaluate a bank. There is unfortunately no mathematical rule to combine all these factors. One has to put all these parameters together and come up with a composite picture of a bank. I will take an example or two in the subsequent posts to evaluate some banks.


The major events over the last few weeks created a small scare for some of the Indian banks. The fear was the level of exposure to the Lehman Brothers bankruptcy and the ongoing credit crisis.
Considering that a lot of us including my family and me have our savings with ICICI bank, I decided to have a look at the Annual report of the bank in detail. Following is the analysis of the bank from a depositor’s perspective (and not an investor’s perspective).
Some positives
The bank raised 20000 Crs in 2007. As a result of the issue, the CAR (capital adequacy ratio) now stands at around 14%. The bank has an annual profit of around 4150 Crs and a consolidated profit (including subsidiaries) of around 3100 Crs.
ICICI has several subsidiaries. In aggregate the subsidiaries are making losses, mainly due to the insurance sub. Due to the insurance accounting (expensing the policy expenses in the current year), the insurance subsidiary has been showing increasing losses as it grows. However the subsidiary has value, which is growing. In all, my personal estimate for the valuation of subsidiaries is around 23000 Crs, which is around 30% of market cap and almost 50% of book value
The bank has been in the limelight due to the losses incurred by the collapse of Lehman brothers. This has been a case of availability bias. The market has been focused on the dramatic instead of the important (as usual).
The derivative related losses (dramatic) incurred by the bank have been to the tune of around 887 Crs which have been charged to the P&L statement (marked to market) and around 203 Crs, which have been charged to reserves.
AS 30 accounting requires mark to market accounting (and P&L pass-through) for certain derivatives and reserves adjustment for others (read AS30 to understand the details).
At the same time there has been a rise in the Gross NPA from 4850 Crs to 8350 crs. This increase is more important for the bank and its valuation as retail assets account for around 60% of the bank’s assets. However the market did not react strongly to this important change as it is hidden in the Balance sheet. The NPA have increased further in the current quarter. In addition the provision are around 55% of Gross NPA. So there is still an exposure of around 3500 Crs, which could hit the P&L in the future.
accounting is pretty complex
Bank accounting and especially derivative accounting is complex. It is very difficult to make out whether the bank is making or losing money on its entire derivatives exposure at any point of time. The bank discloses the total notional exposure which is atleast 1000 times or more of the net exposure. The profit or loss is a multiple of the net exposure. So it is difficult to figure out the profit or loss on the derivative book based on the bank disclosures alone.
In addition mark to market accounting is also misleading. It is equivalent to drawing your personal profit or loss based on change in share prices. If you think a stock is worth 100 rs, and you bought it for 50 rs and the price dropped to 30, how will you account for it ?
Mark to market accounting says, report a loss of 20 now. If the price jump to 70 in the next quarter then reverse this loss and report an ‘income’ of 40. However you may choose to ignore these swings and say I intend to hold the share for next 3 years and believe the market is mispricing the stock in the interim.
So what is the truth ? frankly there is no objective truth. It depends on the specific instrument and circumstances.  Accounting requires being conservative and hence the loss of 20 in the current quarter.
This is the kind of complexity we are dealing with derivatives. The bank may very well have losses on the portfolio or they may right in saying that these are only notional losses as the underlying credits are still intact.
are there solvency issues ?
I think there are no solvency issues for the bank based on the current losses and statements from the bank. The bank has reduced the credit derivatives by almost 800 Mn usd. This does not mean that the bank will not have losses in the future due to derivatives. There is a huge derivatives exposure (notional) on the banks balance sheet.  

As of March 2008, the fair value for the derivatives was positive and for interest swap is midly negative (page 116) , so the bank is not losing money on those derivatives (as of march 2008) . However this value may turn negative in the future.
However the point to remember that the bank is making around almost 1000 crs per quarter on a standalone basis. In addition it has a high capital cushion and assets in the form of subsidiaries. So there is a decent amount of capital cushion to absorb any of these losses. There is always a risk of unknown losses hiding in the balance sheet in the derivative books due to black swan events.  I frankly cannot evaluate and estimate those losses from publicly available documents.
Finally the trump card for the bank is the concept of ‘Too big to fail’. Do you think the Indian government would risk allowing the bank to fail (second largest bank in the country) and jeopardize the financial system?
valuation based on book value ?
I am amazed at the simplistic valuations done by a lot of people and analysts. For ex: ICICI is selling at X times book value and hence it is a buy !! If you read the Annual report, you will realise the complexity of this company. It would be silly to value the bank based on book value alone

The bank has assets (subsidiaries) and risk (derivative exposure) which are quite difficult to estimate (atleast for me). A simple book value based valuation is a foolish way to value this bank.

The minimum analysis to arrive at the final valuation is to value the bank and its subsidiaries. The derivative exposure and other liabilities need to valued separately and the net value should be derived from the difference. Luckily, investing in stocks is not like exams where I will get flunked for not answering a question. I can always pass on the stock.


update: 09-Nov – A great post on the valuation of financial firms and the diffculty of doing so …see here

I have written on banking earlier. You can find my analysis of allahabad bank here. Most of you must be aware of the subprime crisis. I discussed it briefly here.

Banks and financial instutions by their very nature are highly leveraged organizations. So the risk of bankruptcy and losses is higher with banks. Citibank is one of the largest bank in the world and has seen its stock drop by 35% this year. The CEO has just resigned. You can read all about the crisis here.

So what does citibank and the subprime crisis have to do with banking in india. Well a lot … Let me digress and tell you a short story.

The year is 1996 or maybe 1997. I was starting to invest and saw an article on IFCI (I guess you must have already got the hint or must be thinking ….what a Bozo !). Well, the article said that IFCI is a good opportunity as it was near its 52 week low and had a dividend yield of almost 5-6 % (don’t remember the numbers exactly). So thinking that I had found a good opportunity I promptly bought some stock.

Fast forward: 1998-1999. IFCI is a government controlled institution. Politicians look at it as their piggy bank. So if you are a well connected businessman, launch a project, get funding from IFCI, take your money out and refer the company to BIFR. So by 1999, I think IFCI had more than 12% NPA and was bankrupt. There was hardly any dividend and the stock had tanked by more than 70%.

So the moral is …..

1. Don’t base your investments on someone else’s analysis
2. Investing based only on dividend yields is not a good idea. Investing in financial institution based only on dividend yields is a very bad idea unless the financial institution is sound and can maintain the dividend.

So what has happened with citibank is possible with Indian banks too. Banks have a lot of leeway in hiding bad loans. Indian public sector banks due to political interference can end up with even more and these bad loans or assets come out only later. It is difficult to judge asset quality just from the balance sheet

Added note: I have an NRI friend who had invested in citibank based on the dividend yield. Just out of curiosity I downloaded the AR of the bank and my head started spining. It is more than 100 pages, very complex and very difficult to understand (especially for me and may be the CEO too who got fired for not understanding or maybe underestimating the risks).