Banking on Trust-II

In the last post Banking on Trust-1 we discussed Bruce Berkowitz ideas on bank as an earning machine, how it can be an earning machine (by turning into a franchise) and Wells Fargo’s success in the same. I ended the last post by promising we will be discussing Berkowitz’s idea of WFC’s management. But before I go forward I want to make a detour.  I am going to talk about banking business in general and then in the last few paras talk about Berkowitz’s observations. Financial sector is often the least understood sector. Investors and laymen alike find it increasingly difficult to make sense of banking as a business. When I first started this series, I wanted to build my circle of competency in this sector as well as keep a public record of my understanding for others to learn and correct. So without much ado, here it is.

In a business like banking it is important to understand the driving factors in its success.

To explore this idea let us try to understand the problems of its CEO. If we explore this idea, then lots of factors become clearer. Better yet, let us try to form a financial business from scratch. Each case will be taken up as a case study #.

Lets get started:

(Note: In all previous financial statements, the balance sheet preceded the income statement by 1 year)

Case Study #1: We have 1000$ invested in a bond portfolio, which yields 22% annually and matures in 10 years. The loan portfolio is completely riskless. 

In this case, the balance sheet and the income sheet looks like this:

A simple credit financing operation

But note, that we would like to enhance the Return on Equity. The RoE of the operation is not very high. As a result, we would like to employ DuPont methodology for analysing RoE.

It says:

ROE = Profit Margin X Asset Turnover X Leverage

So employing this we understand that, increasing profit margin will either require taking more risk on the loan book or procuring cheaper capital. Both of which is say for the sake of argument not possible. Since the portfolio churns once in 10 years, increasing the asset turnover is also not possible. So the only way is to take in more debt.

Case Study #2: Intaking a debt of $ 500 at the rate of 2% to the existing operation.

As a result the balance sheet and income statement will look like this:

The leverage is mere 1.5 and ROE has also jumped by a similar level. However in two aspects this scenario is divorced from reality.

Consider this, when a financing entity offers me loan at 2% which is risky asset, how can this credit financing operation find riskless assets at 22% yield?

So it will be prudent to recognise, understand and acknowledge that the bond portfolio cannot be riskless and definitely the interest spread cannot be this high.

Analogously, if risk and yield are correlated, then undertaking investments in riskless assets will lead to negative spread. That is cost of funds will exceed the yield from those funds. So let us still for the sake of argument  continue to believe that they are riskless, however reduce the spread from 20% (22%-2%) to 1%.

In other words, the yield on commercial loans we make here is 3%

Case Study #3: The yield on commercial loans is 3%


The ROE has fallen once more! And this reflects a more practical scenario. So is banking business doomed with rock bottom ROE?

No, not at all. Remember , we can pull the leverage couple of notches. Now its mere 1.5x leverage. Lets jack it up!

Lets make it 20:1, no 25:1 leverage.

Case Study#4: Effect of 25:1 leverage on ROE

Note, the increase on ROE. The ROE is now 19.60%.

By now quite a few ideas have started coming to your brain,  I am sure. But lets explore a bit more.

We need to consider the inherent riskiness of the loans as well. So lets do it.

Lets indeed treat the loans as risky assets. Which implies, there is a definitive probability of certain loans going bad. Now as we can foresee, if we leave the entity (financial entity) to run as it is, with no difference made to account for the riskiness of the loans, we can face these problems:

  •  The Net Interest Income(NII) will be extremely volatile.
  • In good times when cost of fund is cheap, the banks will how high profits and in bad times when the default rate rises the profits of banks plunge- making banks extremely coupled with the economy and more risky to own during the downtimes.
  • Excess leverage of the banks make their balance sheets “fragile”- a small amount of default can wipe their equity portion and may even hamper their debt part.

These reasons and more, make it evident that “business as usual” is not wise. We will have to protect ourselves against bad loans while making some kind of provisioning to account for the riskiness of the portfolio.

If we think a bit more deeply, we can say for a bank the level of interest it can charge can be gleaned from this relationship:
Here r_{l} is the interest charged from the debtors, r_{b} is the risk free rate, E(d) is the expected annual default rate, k is the risk premium and c is the operating cost per unit of loan disbursed.

Thus, if L is the loan disbursed on average, the cost of funding is r_{d} and \Delta BL is the bad loans charged off (i.e completely written off), then the NII for the bank in any given year will be:

NII = \tilde{L}*((r_{b}+E(d)+k)-r_{d}) - \Delta \tilde{BL}

Note: Here we are using \tilde{L} and \tilde{BL} to imply that these are not constants but are changing variables. Also we have not used the cost per unit loan in the above equation because we are considering Net Interest Income (i.e post overheads).

This is a situation which considers that there is no provisioning. As a result NII will keep on changing as per the whims of \Delta BL. In bad times when the bad loans increase Net Interest Income will plunge perhaps even going into losses, and in good times when bad loans are below the E(d) rates, Net Interest Income soars.

So remember we are the CEO of this investment operation. It is our duty to see that the health of the business is not adversely affected in any case. We have to thus protect ourselves from the bad loans coming to bite us. Especially in the bad times. Let us create a separate account, where part of the net interest income will be kept aside to cushion the fluctuations in \tilde{BL} (bad loans).

Hence let us keep aside a fixed proportion, \gamma annually of the value at risk.
What is the value at risk here?

Naturally it is expected rate of default E(d) times the loan disbursed L.
Hence, provisioning equals \gamma * E(d)\tilde{L}.
So for each year, we will set up the amount by which provisioning exceeds \Delta BL.

As a result the Net Interest Income will look like this:

NII = \tilde{L} \cdot ((r_{b}+E(d)+k)-r_{d})-\Delta \tilde{BL} - (\gamma \cdot E(d) \cdot \tilde{L} - \Delta \tilde{BL})

On manipulating, we can find something interesting. We find:

NII = \tilde{L} \cdot [r_{b} +k - r_{d}] + (1-\gamma)\cdot \tilde{L} \cdot E(d) if LLP > 0

Here LLP implies Loan Loss Provisions.

In case LLP=0 , the situation reverts back to the representation where there was no cushion.
So what do we see with all this?

We see, the cyclical fluctuations in \Delta BL , bad loans i.e. is completely removed.
We also see, that it is a pseudo-expense.

Just like a manufacturing company can eschew depreciation completely and every few years it can show an extraordinary expense to replace the machinery, similarly without LLP the bank will show extraordinary losses every few years.
By using provisioning (which can be thought of as banking equivalent of depreciation), the CEO are able to smoothen out the profits and cycles

Thus let us account for the riskiness of the assets. Let us consider that expected value of defaults is 10%, and loan amount as per the previous case study was $26,000. Thus the value at risk is $2600. We set aside a proportion of this value at risk say about 1% as provisioning.

Furthermore let us assume there is no charge off this year.

Case Study #5: With 1% provisioning of VaR

Now we are getting closer to our ideas here:

What do we observe here?

1. Prudency demands we should be good at predicting the quality of our loan book, and honest in reflecting the same in our provisioning calculations.

2. In bull runs, there are additional pressure from market, from shareholders to constantly  beat the expectations. And it is that crucial time when CEO has to shut out the noise and allocate provisioning capital liberally. Because when economy contracts the bad loans will have to be offset from this.

3. CEO should build moats around their operation, so that customers dont switch easily at the drop of a hat and some amount of premium pricing can be charged for the services. .

4.  CEOs should be extremely low cost operators, so that the margins can be increased.

In this light it is important to note that the management of Wells Fargo have been prudent in recognizing risks,honest in acknowledging them and cautious in  accounting for them.

So what do we see, when Wells Fargo is hit badly by California Real Estate shock of 1990.

The management errs on the side of caution by high provisioning for the loan losses. So much so that, the earnings power of WFC is completely disguised.  Secondly, a manager who over reserves is one who will understand the price of risk. In other words, it is an indication of cautious, responsible and able managers.

Banking is a business where there is no place for gung ho managers. Being a business so fragile a tiny amount of neglect towards risk can wipe out the bank.

Berkowitz at one point drives home  harder the fact that Wells Fargo has captured the markets so heavily, mark my words, he says if Wells Fargo is to close down tomorrow then there would be significant shortfalls.

That, my friend is the mark of a company which has locked out any competition on its turf. In short it has got a moat around it!

What does it imply?

One thing. Management is the best as they come.

Wells Fargo is also the most efficient bank out there and has an ROE of 20% (as of 1990). Bruce Berkowitz at this point introduces PTPP Earnings, that is core earnings for a bank. It is called pre tax pre provisioning earnings. He points out that PTPP earnings of WFC stands at an astonishing $33.

And their provisioning which is extremely conservative, and assumes a complete charge off is not so. In fact their non performing assets are earning  a modest amount of interest.

Buffett reflects Berkowitz idea toto. He echoes that management of a bank is the singlemost make or break factor. The next important factor is costs. Wells Fargo has an average cost of deposits at 0.36%.

Similarly its return on assets stand at 1.43%, return on equity at 13% and return on tangible equity stand at 18.3%. Compared to any cheap community banks Wells Fargo is far better in quality and far better in value.

This post has stretched far beyond usual ones. For now taking your leave. Adios.

In the next post of Banking on Trust series, I will be discussing the business aspects of Wells Fargo- its competitive advantages etc.




One thought on “Banking on Trust-II

  1. Pingback: Banking on Trust-III | Meditations

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s