Interview with Prof.

Sanjay Bakshi

Chetan Parikh interviewed Prof. Sanjay Bakshi, CEO of Tactica Capital Management on 22nd
March 2007 in Gurgaon.

CIO: Thank you for giving us the time to talk to you.

Prof. Sanjay Bakshi: It’s a pleasure to meet you here in my office. We are meeting after more
than four years and I remember very fondly what happened on the previous occasion! Indeed
I still get fan mails from people about the talk I delivered at your invitation at the Oxford
Bookstore in 2002. I don’t talk to the media. I talk to my students in the classroom at MDI (six
months in a year) and I frequently talk to some value-oriented friends. You are the only one
whom I have met from the media in the last four years!

CIO: Many thanks.

Prof. Sanjay Bakshi: So it’s a pleasure to have you here and I just want to open up the
presentation, which I had given to the invitees at Oxford Bookstore in 2002. This presentation
was titled, “Value Investing – a Conservative Way to Invest.” I had started by describing what
is value investing and why the focus of conservative investor is first one not losing money
than on making it (downside risk more important than upside potential) and then I gone on to
the better part of the presentation, which essentially dealt with three value investing themes.
These were: (1) Cash bargains; (2) Debt-capacity bargains; (3) Debt pay-down.

I only focused on these three themes in that presentation and gave a number of examples.
Some were past examples whereas some others were current examples at the time of that
talk. With the benefit of hindsight, I can now tell you that things have worked out pretty well
for me professionally using those three themes.

But knowledge is incremental, especially in the profession of security analysis, and

experience is a good teacher � so I have evolved over the years and there have been many
changes in the way I think about investing. In my early years I was very influenced by Mr.
Buffett. But over the last few years, I have become more influenced by Mr. Graham.
Hopefully, today we will interpret Graham�s bible for this profession, the Security Analysis

www.capitalideasonline.com
Page - 1
 Interview with Prof. Sanjay Bakshi

and of course the Intelligent Investor.

Ten Deep Value Themes

So today, I am going to continue with where I left off in 2002 and focus on ten deep value
themes. Some of these are old themes and I just want to revisit them. And then I want to talk
about other themes on which I have accumulated some wisdom over the years. Specifically, I
am going to talk about:

The changes in my thought process over the years

The classic Ben Graham themes

Capital structure related themes

FM strategies � I won�t reveal what �FM� stands for right now – a bit of mystery I want to
introduce to hopefully keep your readers awake!

Dividend policy related themes

Event-driven themes

Availability bias related themes � after my last interaction with you I became hugely
interested in psychology and spent a lot of my time learning it. Indeed, I now teach a paper
on behavioral finance which focuses on the �foolish man models� from psychology rather
than �the rational man models� from economics which I had been mis(taught). Over the
years, I have developed some themes which arise purely out of biases most people and
certainly Mr. Market suffer from. And some of these are hugely important in my view. One of
them, of course, is the availability bias. And I have a whole set of strategies coming out of
that bias.

Mean reversion strategies

Value-plus-momentum strategies, and

Over-optimism related strategies. The last one is essentially to do with shorting overvalued

www.capitalideasonline.com
Page - 2
 Interview with Prof. Sanjay Bakshi

stocks as a hedge against a major decline the the prices of deep-value shares.

1) The Classic Ben Graham themes:

Let’s just go to classic Ben Graham strategy. Of which six of them are:

i. Cash bargains

ii. Debt capacity bargains

iii. Earning yield bargains

iv. Large, unpopular companies

v. Low-priced common stock, and

vi. Special situations

The subject is vast. Graham’s book (Security Analysis) is very useful. And if one really applies
in a creative way the essence of what Graham taught in that book, to what is happening
around us now, one can, in my view, build an entire life�s career around just a few of the
things he taught. Personally, for me it’s been very interesting to see how the same logic
works even seventy years after the book was first published. And Graham put it down his
ideas so eloquently and they are all there and people just don’t use those ideas in the way I
think they should be used.

i. Cash bargains

I talked about cash bargains in 2002. We know that, if the market value of a company is less
than the net cash (cash and marketable securities net of current liabilities and debt) in its
possession, then you�re getting the fixed assets and assets other than cash, essentially for
free. But Graham gave a warning �Be careful about cash being dissipated away if there is a
loss-making business out there.� So you may have a company whose stock sells at a price
making it a �cash bargain� and the business is losing money. In such cases, the market is
right in valuing the stock at below cash, because the investors will not see the cash. And

www.capitalideasonline.com
Page - 3
 Interview with Prof. Sanjay Bakshi

that’s what I too had said in 2002, but then I had some additional thoughts on that subject,
over the years.

Take holding companies, for example Nalwa Sons or Jindal South West Holdings or
Consolidated Finvest, all of which are so-called-cash bargains. My simple question here is
where is the catalyst? Is it a family dispute? Is it the presence of Mr. Soros, for example in the
case of Jindal Southwest Holdings? Your know Jindal South West at Rs 225 at present is really
interesting because the company holds shares in JSW Steel on which futures and options can
now be traded. So one could technically go long in the holding company and short the
underlying and have a very interesting trade out there. But you can only make money on the
trade if you narrow the spread. Alternatively, one could just buy the holding company as a
very cheap way of getting an interest in JSW Steel. Well whether Mr. Soros will be able to
narrow the spread or whether people who are in that particular situation will be able to do it
and whether it make sense for us to buy into that situation or not, that’s debatable.

But in the absence of a catalyst, I don’t feel very excited about holding companies as I used
to at one point of time. I now feel that it takes a huge amount of patience for such situations
to work out in the absence of a catalyst. Moreover, the basic structure of holding companies,
as Graham had mentioned in his book, is a very defective corporate structure. It may be a
wonderful structure from the controlling stockholders� point of view – from the market’s
point of view it’s a very defective structure. The more layers you put in between the eventual
property and the ultimate owners, the more reluctant would the market be to give full value
to that property.

This is pretty much the case with close-ended mutual funds and conglomerates, which are
the other two forms of bad corporate structures. Holding companies are a bad corporate
structure from minority shareholders� viewpoint. Indeed, in my view, holding companies are
even more inferior to close-ended mutual funds because at least the latter have a limited life.
As per the Indian regulations, all closed-ended funds are to be liquidated by the date
mentioned in the initial offer document, which ensures that the discount to NAV will

www.capitalideasonline.com
Page - 4
 Interview with Prof. Sanjay Bakshi

eventually vanish otherwise it would violate the no-arbitrage principle in financial markets
(None of the funds are ultimately liquidated � all of them are converted into open-ended
funds thus demonstrating the classic agency conflict involved here but that�s another story).
Unlike closed-ended mutual funds, however, holding companies have an unlimited life
because they are companies and not funds. This makes holding company akin to perpetual
close-ended funds run by managements who really have no interest in unlocking value for
their shareholders. Its no surprise that discounts in holding companies are much more than in
close-ended mutual funds like Morgan Stanley Growth Fund.

Operating companies with cash

Another thing I want to say about cash bargains is that operating companies with cash are far
better than �boxes of cash� i.e. holding companies with no operating business. So, in
contrast to pure holding companies, which are nothing but �boxes of cash�, if you have an
operating business which generates surplus cash, and in addition you have a substantial cash
on the balance sheet, and if the company is under-leveraged, and if the stock price is not
presently implying a cash bargain in the pure Grahamian sense, but is low enough, so that
one can actually estimate as to how soon will this company become a cash bargain � well
that�s a very attractive combination in my view. It’s far better in terms of relative
attractiveness than just a box of cash because the box of cash is really something they can
take away. They can replace it with some loan and the loan terms could be very onerous. So,
all sorts of things can happen which could prevent a minority investor to be able to realize
that value of surplus cash.

In contrast, when you get a cash-generating operating business, a zero-debt company, lots of
cash on the balance sheet, and a market price of the stock which is not very far from the net
cash on the balance sheet, you have an attractive, low risk, high reward situation. By high
reward, I do not mean a multi-bagger, but rather a return which is at least twice of the AAA
bond yield, which incidentally, is what a Grahamite tries to look for.

So, I have sorted of moved away from pure holding companies without catalysts to

www.capitalideasonline.com
Page - 5
 Interview with Prof. Sanjay Bakshi

companies which are debt-free, selling at low price/earnings multiples – so low that if you
remove the net cash from the market value, the operating business turns out to be selling for
a very low multiple of the underlying corporate earning power. Indeed, you can actually
predict, under reasonable assumptions, as to how many quarters or years will it take for this
company to go below cash if the stock price remained unchanged. And, in my view, that�s a
very useful way of thinking. I think these companies will not sell below cash for a prolonged
period. Unlike boxes of cash, many of which deserve to sell below cash, cash-generative
operating businesses don’t deserve to sell below cash.

So companies like Neyveli Lignite or GAIL or some other companies which can easily be found
would probably fit this criteria One could predict perhaps how many quarters, how many
months, how many years will it take, if the market value were to remain unchanged, for this
company to sell at below cash because you can estimate how much cash is being made and
you can estimate how much dividend is being paid out. Cash is coming in and there are no
other uses of cash. So it�s all there in the cash flow statement and in the earnings
statement and one can make a reasonable judgment as to how many years will it take to �
so if something is just two or three years down the road to end up being below cash, I want to
be an owner of that particular stock. I think it’s a very conservative way of investing,
particularly if you combine it with dividend yield which is fairly satisfactory to begin with. So,
that was my other observation about cash bargains.

Why do markets hate the most easily valued asset on the balance sheet?

CIO: Why do markets hate the most easily valued asset on the balance sheet? Cash is after
all easiest to value. It’s there, but the markets just hate it. Why does that happen?

Prof. Sanjay Bakshi: Well, I think one reason why that happens is that Mr. Market loves EPS.
And cash doesn�t contributes much to that EPS. It produces low treasury income and it
doesn’t show up much in the EPS.

Another reason is that Mr. Market fears that the cash can easily be transported away by the

www.capitalideasonline.com
Page - 6
 Interview with Prof. Sanjay Bakshi

insiders. Now, you can’t transport the plant and machinery, but you can easily siphon off
cash, so to speak. So, it’s not that tough to remove cash and markets are very skeptical and
that skepticism gets reflected in low price/earnings ratios of businesses which has surplus
cash. And that skepticism may often be overdone, by and large the market is not totally
wrong on this. While many Indian companies have been good allocators of capital, the
dividend policy of many many others are pathetic. They should dramatically increase their
payout policies. You see much better values and the cash on the balance sheet would get
much better value in the market. So one of the things which we will come to later when I’m
talking about dividend policies is that what investors can do is to make a point with the
management to increase the dividend payout when the money inside the firm is not being
valued properly by the markets. It is more valuable in the pockets of the investors than in the
pockets of the company.

ii. Debt capacity bargains

On debt capacity bargains, there�s been no change in my views over the last few years.
Graham was right when he wrote that an equity share representing the entire business
cannot be less safe and less valuable than a bond having a claim to only a part thereof. And I
have used that theme over and over again to identify, and, indeed arithmetically prove the
cheapness of some stocks. Mr. Munger taught us that if we have a problem to solve, then if
we reduce the problem to a discipline, which is more fundamental to our own discipline, then
we have a very good basis of solving that problem. Graham�s approach to debt capacity
bargains reduced the equity valuation problem to basic math, a fundamental discipline. So,
when you come down to that level of accuracy, it’s impossible indeed to argue that the debt-
capacity bargain stock you are looking at isn�t cheap.

The private equity boom in this context is important because debt-capacity bargains are
natural candidates for going-private transactions. In my view, the private equity boom is here
to stay. Some people think it’s a bubble and assume that private equity players will stay for a
while and then will go away. However, my view is that as an alternate to the stock market,

www.capitalideasonline.com
Page - 7
 Interview with Prof. Sanjay Bakshi

companies that are not being valued properly are likely to be taken out of the market by
smart private equity players. And that’s generally a good thing. Because if markets are not
giving good and appropriate value to listed stocks and somebody else is willing to pay a
premium over the market value to the controlling and minority stockholders, then he has the
right to enjoy the difference between his estimate of value and what he paid to take the
company private, to the exclusion of the others. And I think that part of the return in private
equity is a very substantial part, plus there is the incentive effect to correct misallocation of
capital from such companies the system. And, of course, the private owners have the
valuable option to decide when to bring the company back to the public markets, which not
co-incidentally, will be when markets are salivating for more stocks in that particular industry,
so that much of extra return is also there. So I think that the private-equity wave is going to
become a much bigger wave. I don’t think that’s a fad. It’s here to stay.

The debt capacity bargains idea, which made me lots of money, led me to think about capital
structure as such. You know how important capital structure is and of course there is plenty
much to learn in Graham’s books on that subject.

CIO: Could you comment on some such companies?

Capital Structure Strategies:-

Prof. Sanjay Bakshi: Yeah. If you look at Abbott India, if you look at the surplus cash in the
company�s possession and if you net it out from the market value of the company, you end
up with a business value which is fairly low, given the quality of the business, given the
returns that the business earns on the capital. So it’s a cheap stock out there. And if it is a
cheap stock, then insiders have strong incentives to take the company private. They are not
doing it (apart from two small buybacks). There are different reasons for that. But these
companies really don’t deserve to be in the market. They should be taken out from the
market. Persistently, they sell at below what they are really worth. And I think a lot of that
has to do with the amount of cash in the balance sheet, and management�s reluctance to
give it back to the company�s owners.

www.capitalideasonline.com
Page - 8
 Interview with Prof. Sanjay Bakshi

In many of such situations, insiders are not really worried about increasing their stake in the
company or increasing the stake of the minority investors by virtue of a buyback or a
leveraged recapitalization which. They are not doing any of those things, but I think they
should. They should really think about it. And I don’t know whether it will happen or not but it
should happen.

CIO: So would you look at Abbott India as a good investment at this level given that it is
cheap but that there may not be any catalyst?

Prof. Sanjay Bakshi: It�s a cheap stock. But if I were to look at it, I would look at it with the
point of view of being a catalyst in the process of getting the company taken private. So, if
they are holding a percentage, why not go and make an open offer for the rest of the
company and become a private owner with the management, which is generally a very good
management. And of course you know the other thing about what would happen if you were
to make such an offer. If somebody were to come out and say well this company is out there
and it’s undervalued and the management is not doing anything about it and if I make a
public offer to buy out and take this company private, how will the other side respond what
Mr Munger calls �effects of effects�)? I think they’ll probably respond by delisting the
company – you have to spur them to do it. They won’t do it on their own but if you nudge
them a bit they will in all likelihood do it. If you have the capital and if you have the
conviction then the odds of losing money are very low.

What happens to a company that has too much equity and what happens when the company
has too much debt?

Modern corporate finance proclaims that capital structures don’t matter. And I think that�s
nonsense. I think it’s better to ignore what Modigliani and Miller has said and instead listen to
what Graham had said on he subject and and what KKR has done. What did Graham say? The
essence of what Graham said, although he never used the phrase �optimal capital
structure� is that there is such a thing as an optimal capital structure. And he used his own
system of credit analysis which is very unique. For example, he never believed in credit

www.capitalideasonline.com
Page - 9
 Interview with Prof. Sanjay Bakshi

ratings because credit raters rate instruments whereas he rated companies. He had a single
rating for a company. In Graham�s framework, if a company was credit worthy and all of its
debt instruments were also credit worthy, otherwise he would not touch it. And he said that if
a company is debt free, then the optimal amount of debt it should have on its balance sheet
is the amount, which would classify as a high-grade investment from the bond-holders�
viewpoint using Graham�s framework of credit analysis.

So in the sense that’s what he meant when he said there is a thing as an optimal capital
structure and there are all sorts of businesses which are not amenable to debt financing, so
they shouldn’t have any debt in them, for example, new startups, or companies having
businesses with very volatile cash flows, but if there is a large company with very low debt-
levels having very stable cash flows then it can have a cheaper source of capital than equity
on its balance sheet. How much debt to put? Well, he had a very simple formula and I think
that works. You don’t have to be exactly right, you can be just roughly right.

And KKR took it to the extreme by putting a huge amount of debt on target companies to
finance their acquisition at large premiums to pre-acquistion market prices. KKR did it and
the private equity boom is showing that it can be done and companies are being taken
private and they are putting debt on the balance sheet. And the main thing that works here is
the change in the incentives. The interesting thing that demolishes the MM argument is that
MM forgot or they didn’t think about the possibility of a jump in EBITDA itself.

KKR showed that by putting large amounts of debt on the balance sheet of an under-leveraed
company you can change the management�s incentives in a manner so that it has much
less desire to spend money on those fancy carpets or the air planes, then certainly the
margins went up and the EBITDA went up and if EBITDA goes up, the enterprise value goes
up. If there is no change in EBITDA, you have more debt, equity will shrink, if you have more
equity, debt will shrink � that�s what MM thought would happen � they forgot that people
respond to incentivres. The total value of the firm is independent of capital structure if you
ignore, among other things, the incentive effects of debt on a mismanaged, under-leveraged

www.capitalideasonline.com
Page - 10
 Interview with Prof. Sanjay Bakshi

company. But, if EBITDA itself changes, the value goes up and that value will go to the equity
because the claimants to the debt are fixed. Incidentally, there is a wonderful Harvard case
on Sealed Air, which I think every investor should read.

So, all of this, led me to think about what happens when a company has too much equity and
what happens when the company has too much debt. So that sort of morphed into different
things evolving from the idea of capital structure. So if the company has too much equity and
the Abbott kind of situation or the stock is a debt-capacity bargain, there is too much cash,
they need to distribute the cash. And there have been some minor transactions where we
have made some money like Hindustan Lever shareholders got bonus debentures and
Thermax shareholders got bonus preference shares. So those were transactions, which were
effectively distribution of cash by changing eh capital structure. They were delivering
shareholders instruments, which were debt instruments that they could sell in the market
priced as debt instruments, but they were changing the capital structure. They were moving
towards what was optimal and it was done at a time when the companies were undervalued
so there was a major upside returns for investors who understand how these things work. I
think this model is replicable by other similar companies in India � there are lots of such
companies.

On the flip side you have companies with restrictive capital structures. There is too much
debt and too little equity. There is money to be made there too because the fact that you are
observing the company now which has too much debt means that you were not participating
in the original calamity which caused all the debt to come up in the first place. So the market
is putting a very low value on the enterprise, because it’s highly leveraged and it sells at a
ridiculously low PE multiple. But, if the present value of debt was reduced as opposed to the
book value of debt,, either voluntarily or otherwise under CDR (Corporate Debt
Restructuring), which has been a major trigger for such transactions to happen, you are on to
something. So we look very closely at corporate debt restructuring as the theme arising out
of the debt reduction theme, arising out of the capital structure theme.

www.capitalideasonline.com
Page - 11
 Interview with Prof. Sanjay Bakshi

Equity is just a balancing figure and if you can value the corporate entity without thinking of
the capital structure and net off what you think that debt is worth under changed
circumstances, there may be instanes where you have a hugely under- valued equity.. Of
course, we have to project as to how that value will rise as debt is reduced and how likely is it
that debt will get reduced and whether their source of debt reduction would be voluntary or
involuntary. It could be probably voluntary in CDR cases under a strict timetable. They have
to adhere to the timetable. The interest of the minority investors in such transactions get
perfectly aligned with the interest of the lenders to the firm because under the CDR package,
the lenders will get the company�s management to stop all that stupid diversification and
they have people to ensure that it happens and they stop the dividend, and instead focus on
debt service and debt reduction, which is exactly what a value investor wants. Paradoxically,
the skipping of the dividend produces a very low market price. So you have a system now
where you can effectively project as to when will this company be substantially less
leveraged which means an automatic expansion of the equity valuation and when will the
dividend come. Because there are conditions laid down and the CDR package is there on the
side and one can read the model agreements and what kind of restrictions lenders put on the
whole process and you can effectively predict roughly as to when this company will come
back on the dividend list and we know that the stock prices are extremely sensitive to the
announcement of dividends from companies which have skipped dividends for a number of
years.

So, I think that’s a very interesting field in itself. And there have been so many cases of CDR
and you know I don’t think any deal that passes through that desk should escape the notice
of a value investor.

iii. Earnings yield bargains

Earning yield bargains are a classic case of inverting the PE multiple which Graham did and
he showed that if you can find stocks that promise to earn at least twice of AAA bond yield on
your opurchase price, then you should have it in a diversified portfolio.. That simple, elegant

www.capitalideasonline.com
Page - 12
 Interview with Prof. Sanjay Bakshi

idea still works in my view. The idea of earnings yield being stated as at least twice of AAA
bond is a very elegant idea because it removes the need to think about cost to capital. And I
want to relate this to Herb Simon�s model of satisfysing as opposed to Markowitz� theory
on optimizing. You know all the optimizing models and they talk about the capital pricing
models and the betas and so on and try to estimate exactly what the cost of capital is. If an
investor were to think about allocating capital in opportunities where you can get twice of
AAA bond yield, then he is automatically in a system where if the interest rates were to go
up, the effective PE multiple he would be allowed to pay for a stock would go down.

And that works, it’s simple and it’s a very elegant way of thinking- and not just about stocks
but about life too as in if you want to achieve happiness then sometimes the best approach is
to lower your expectations i.e. be a satisficer and not an optimizer � have a predetermined,
and adjustable aspiration level (twice of AAA bond yield in investing), and stop searching for
greener pastures once you find one or many, which meet your aspiration level. So I won’t say
much here because this is very simple stuff and it’s been discussed again and again and is
one of the chief criteria pertaining to Ben Graham’s stock selection for purchase.

iv. Large unpopular companies

Graham also talked about the large, unpopular company as a deep value investing theme,
and that of course works a lot because large companies become unpopular for all sorts of
wrong reasons. We have a theme called �dogs of nifty�, statistically we measure which of
those nifty stocks are the cheapest ones at any given point of time. So it makes sense to look
at these companies exactly for the reason that Graham envisioned. These companies will go
back much faster than the small and popular companies. So to that extent if you have two
situations; a large and unpopular company (unpopular for the wrong reasons) and a small
and unpopular company (again unpopular for the wrong reasons), then you should obviously
choose the larger one even if the margin of safety is bit smaller than in the case of the
smaller company. Even though the rate of return on investment could be smaller in the large
one, the probability of the return coming to you rather quickly is that much higher. So it

www.capitalideasonline.com
Page - 13
 Interview with Prof. Sanjay Bakshi

makes sense to go for the large ones even when the value-price gap is smaller.

But large unpopular companies — if you buy them when they’re cheap and they are
unpopular for the wrong reasons, which means they are likely to go back to their normal
valuations, should be seen as a subset of a broader universe of mean-reversion strategies
which is an area where we have done some work and we are continuing to develop more
ideas over there. I will come to that later. I think that’s a very fundamental model. The
reversion of the mean is a mental model from statistics � its a very powerful model and
there are a number of strategies that come out of it.

So, its very natural for us to look at deeply out-of-favour sectors because our experience tells
us that they go out of favor often times for the wrong reasons. One sector which has been
coming up rather frequently on my screen is sugar. It’s deeply out of favor and the outlook is
very terrible, which is exactly what we are looking for. The near term outlook has to look
terrible otherwise how else will it be out of favor in the first place.? So, I think there will be
money made in sugar stocks eventually and it may be early days but one would get paid for
holding on if one picks the right stocks. I�m still working on this. So I don’t want to talk much
more about sugar right now.

v. Low-Priced common stocks

Mr. Graham also talked about low-priced common stocks and I have here an extract from
Security Analysis when he talked about the arithmetical advantage of a low absolute price. A
stock may move from 10 to 15 much faster than it would move from 1,000 to 1,500 and that
part of the arithmetical advantage is there. But that doesn’t mean that every stock, which is
at a low absolute price, is a cheap stock. And he talked about super low prices where there
are billions and billions of shares outstanding and the market cap is so high in relation to the
revenues and the earnings and the asset values that what you�d think is a low-price is
actually a pseudo low price and he has given some examples of that. So he talked about PSR
(price-sales-ratio). PSR is a very important ratio and I think that’s fundamentally important in
the context of low-priced common stocks, that the market value of the shares of the firm

www.capitalideasonline.com
Page - 14
 Interview with Prof. Sanjay Bakshi

should be very small in relation to the annual revenues of the firm.

And here I think I can talk to you about a company, which we have been recently looking at
Nagarjuna Fertilizers – the stock has come down from Rs.21 to Rs.12 in the recent decline in
the market valuations.

The company is hugely cash generative. The depreciation expense on the books is much
higher than the capex required. Over the last several years they have decided to go into the
refinery business about which I don’t think they have any clue about but there are many
ways in which you can look at that problem.

The way I look at it is that well, they are not going to put any more money. They are not
allowed to put any more money. They have already put all the money that is required to be
put in that refinery and how do you value that part of the business? And if you put a zero
value to that business then how does it look now? So we start with a zero value from the
refinery and see whether it looks attractive. It looks attractive precisely because 1) no more
money is likely to be thrown in that direction 2) the company is under CDR and they are
required under the strict timetable to reduce debt and I am informed by the company that
they are adhering to that and they are indeed ahead of the schedule specified in the CDR
scheme.

And in a stock which is a highly leveraged situation- they have Rs 1,400 crores of debt on a
Rs.500 crores market cap roughly speaking and a Rs 12 stock price and 42 crores shares
outstanding, so even if they reduce debt by Rs.300 or 400 crores which they can easily do in
a single year, if they really wanted to, the upside for the equity is very significant. And you
also have a possibility of a restoration of dividend, which would be a very interesting catalyst.
So that’s the kind of things we�ll want to look at when a low price common stock comes
about the screen because we have a low PSR, we have a highly leveraged balance sheet, we
have a low absolute stock price and there are multiple triggers out there.

But if you were to value the company based on EPS and PE multiples, they look expensive

www.capitalideasonline.com
Page - 15
 Interview with Prof. Sanjay Bakshi

because the earnings are not there. The earnings are not there because the depreciation
expenses are so high but if you look at the cash flow statement, you will easily get different,
and right conclusions.

vi. Special situations

Mr. Graham talked about special situations and I don’t want to talk much about them
because it’s a huge subject. Initially, I started my career in risk arbitrage, or special situations
as Mr. Graham called them. Some people call them event-driven strategies. We do a lot of
work in that area and there has been an explosion in the volume of corporate events, which
provide us with opportunities in this area and that trend is going to remain. And we get
access to these important announcements from the exchanges and that�s our starting point.
We don’t respond to anticipated events as Mr. Buffett recommends. One can lose enough
money on announced events. So we tend to ignore events anticipated by media and the
analysts. It�s been a great adventure to use special situations more as a way of thinking
because it forces you to think probabilistically just as Robert Rubin describes in his book, �In
an Uncertain World� as well as, in his various talks. It’s the purest form of security analysis,
in my view, and even Mr. Graham almost said that.

A few things on special situations that I have learnt along the way

If you’re working on transactions, which involve say, a de-merger, and they have announced
it and there are milestones and there is the perceived value-price gap in your view. This gap
is not going to be narrowed gradually. Special situations do not behave like zero coupon
bonds approaching maturity. As milestones are achieved, as various approvals come in, the
shareholders say yes, the creditors say yes, the courts say yes – they go through various
levels – a huge part of the return comes towards the end when a lot of the milestones have
been crossed. And that, of course, reminds me of Kelly. The Kelly criteria is fundamentally
important here which means that as the probability of return goes up, you should increase
the amount of money behind a particular transaction. So we have been doing that a lot. One
of the few things I have learnt over the last few years is how to use the essence of the

www.capitalideasonline.com
Page - 16
 Interview with Prof. Sanjay Bakshi

Kelly�s idea – the elementary concept of increasing the weights, if the odds are moving in
your favor. So, the market in that sense is inefficient because it gives you the opportunity to
make money in a transaction on multiple occasions over its life. It does require a lot of
intensive interaction, reading legal documents, talking to the company secretary, getting all
sorts of data, interpreting the data and evaluating whether a setback is a permanent setback
or is it just a blip, and in a deal when you are looking at a transaction like a spin-off, or its a
merger-arbitrage deal, there are always surprises and most surprises are bad surprises. But
in the end it tends to work out in most cases. If you know that in the end it is going to work
out, then you will get a lot of chances during the course of this deal to buy in.

What really happens is when people announce events (in-fact I think a bit of information with
some people who know about the impending announcement buy the stock in advance) there
is a lot of hype and excitement created. But then it sort of cools off it tires many people to
remain invested in a deal which is taking much longer than they originally anticipated and in
the meantime all sorts of lovely things are happening outside the deal which people are
missing so there is temptation to sell out on the sign of first adversity, or negative surprise.

But I think that�s the best time to get in � when there is the first sign of some adversity,
something going wrong with the deal, some delay, some court order, some dispute and if you
try out such transactions, you will see there is plenty of volatility coming out of the event risk
itself. So the event risk is fundamental here to understand and the classic case my mind
comes to over here, we did pretty well over the GE Shipping de-merger recently and there
was a huge amount of volatility involved in that as you know and the deal almost didn’t
happen. In fact it was canceled and it then it came back. So one of the things that you learn
when you do such a situation is that it’s not done until it’s done. That’s very important to
understand.

Things were not really immune from market risks

The other thing which surprised me, which I learned in the May 2006 crash, was that these
things were not really immune from market risks. You know they appear to be immune but

www.capitalideasonline.com
Page - 17
 Interview with Prof. Sanjay Bakshi

they are not immune because when the market goes out for a toss, there is a huge crash.
Then everything goes down including special situations because of reasons to do with
liquidity and you know investors need money and they are forced to sell and there are
margin traders who are selling out and the spread widens to even these special situations.
So, if you are a special situation analyst you should be careful that you are not leveraged to
the point where you could get hit because you will get hit, if there is a systemic risk out there
in market and that is something which I learned and I paid a price for it. So I am talking from
a standpoint of student who has paid a very large tuition fee on that lesson. So you’re not
really immune or substantially immune unless you have taken out the risk totally by hedging,
using futures or options � and that is not always possible because the NSE shuts down
trading in futures in many such transactions prior to the consummation of the deal.

FM strategies

FM strategies � �FM� stands for Frequency Magnitude. That’s the new thing that I have
worked on over the last few years and I love that content specially coming out of Kelly and
evolving it. I learnt it from Nassim Taleb (although the term �frequency-magnitude� was
first seen by me in a paper by Michael Mauboussin), and obviously you would know what he
was talking about in his book. Well he said that there is a 70% chance the market will go up
and there is a 30% chance the market will go down, even so, he was short the market. Why
was he doing that? Well, the problem is that most people focus only on frequencies. They
don’t look at the magnitudes and the expected values, which are arrived at by multiplying
frequencies and magnitudes of various scenarios and then adding them up.

And in this case- as you can see in the table on the screen- the expected value of is -2.3%.
So, on an expected-value basis, it makes sense to be short the market even though you
expect that it will go up 70% of the time and 30% of chance that it will go down.

The human brain, as Nassim Taleb says, is not wired to think in expected-value terms. It is a
counterintuitive way to think and its basic Fermat Pascal, which Mr. Munger talks of. And, of
course, Mr. Buffett too said that one has to �take the probability times the amount of

www.capitalideasonline.com
Page - 18
 Interview with Prof. Sanjay Bakshi

possible loss from the probability of gain times the amount of possible gain, that is what we
are trying to do. It�s imperfect but that’s what it’s all about.� Thinking in terms frequency
magnitude really works. It�s the basic way of thinking and that’s exactly what Robert Rubin
has been saying in when he asks us, through his wonderful book and speeches to do
probabilistic thinking.

CIO: Even Mohnish Pabrai talks about low down side plus high uncertainties equaling high
return possibilities.

Prof. Sanjay Bakshi: That’s the same thing but said elegantly. So we have done some work in
these areas and we find there are plenty of opportunities to think in that particular way and
you have to train your mind to think like that.

For example, just to illustrate the idea, suppose that we get to know of a company, which has
announced its intention to pay a dividend for the first time, or after a long gap. We know that
in most value stocks, market prices are extremely sensitive to dividend policy, however in
this case the stock has not yet reacted. So it might make sense to buy that stock. You have
to think about the chances of paying the dividend, the likely quantum of it, the likely outcome
in the market when the dividend is announced, and just before the stock turns ex-dividend,
and also the outcome when dividend is not announced after all (yes it can and does happen).
And oftentimes, you will conclude that the stock should be bought because if you are wrong
you will be getting out at a very small loss or no loss and if you are right, there is a lot of
money to be made.

Similarly, the excise duty on some product might be reduced. So there might be an upside for
a certain industry, but market has not yet sensed it. So we just buy it with the expectation
that if it does happen it will be good, if it doesn’t happen we can get out without much of a
loss. It’s not fully risk-free because the market risk is still there, but its better way of doing
things than just buying on the gut.

So thinking by using FM has vast practical applications in security analysis – it is

www.capitalideasonline.com
Page - 19
 Interview with Prof. Sanjay Bakshi

mathematical, it is analytical, and it is amenable to analysis in the same manner as when we

analyze the companies. So, we do try to keep track of market sensitive announcements,
which have not yet been factored in by the market and use FM to exploit some obvious
inefficiency in the market.

Trade-offs

Keeping in mind the overall risk control and how much money should be allocated to a given
opportunity, oftentimes we come to the conclusion that the expected returns are good but
there are other things that are happening which are more interesting � the opportunity cost
mental model. You know if you have a chance to deploy your money at 20% per annum for a
long time and if you had a chance to make 3% or 4% return over a month, I would take the
first any day. So there are two kinds of trade-offs I have to think about. I think it should also
be thought of in just general value perspective. Why do value stocks become amenable to
frequency magnitude kind of a framework? I think the answer lies in the way news comes,
and in the way markets respond to the news. Well there is asymmetry here, which is very
important to understand.

In deep value stocks when some important news comes the market�s response is
asymmetric depending on whether the news I good or bad. The stock is cheap because it�s
out of favor. So the market doesn’t expect much from the company anyways. Now if you
assume the news that is coming in about the company is random, it could be good or it could
be bad, we don’t know. It’s just toss of a coin. In fact if you think about mean reversion, it
should not be random. It should be more likely to be good because in mean reversion, good
things are more likely to happen to companies which are deeply out of favor for a fairly long
time than bad things. But even if you ignore that and if you assume its random, then
negative news will not have much of an impact on the market price but even minor positive
news will have a significant upside potential for you. And that�s why FM way of thinking
really helps.

It�s a classic way of thinking. For example, when the quarterly results are going to come

www.capitalideasonline.com
Page - 20
 Interview with Prof. Sanjay Bakshi

out, are they going to surprise the market? More chances of a positive surprise than a
negative surprise means that you have a system where you can take an in a deep-value
stock with the expectation of a positive surprise with all the lovely magnitude of that
outcome factored in your thinking along with the magnitude of the outcome of the absence of
a positive surprise.

So, as a subset of a value portfolio, FM makes immense sense. What really matters is not
whether the news is good or bad, but whether it will surprised the market and you have more
likely good surprises than bad surprises in deep value stocks, over time. So what you get is
asymmetric payoffs with favorable odds and that’s what essentially you are looking for.

Dividend policy related strategies

Back to Modigliani & Miller – Was it the real world? I think dividend policy matters a great
deal despite what two Nobel laureates said. They said that, �It doesn’t matter at all.�

Modigliani & Miller assumed that capital markets would always welcome companies with
open arms, if they were to overpay dividends and now needed new capital injection to fund
their investment plans. The whole elegance of the idea of the irrelevance of dividend was
that, �if they pay more, so what? If the company over pays, it can always raise back more
equity capital.� Now there are some implied assumptions in that. One of the implied
assumptions is that the markets are efficient, the value of the firm is equal to the market
price of the firm and therefore if the company does over pay, it takes back new capital which
is issued at firm value, so there is no dilution risk for old owners. The reality is very different.
They also assumed that managements will have zero incentives to squander the cash than to
pay it out to investors i.e. there are no agency problems. They ignored the wisdom of what
Peter Lynch’s called as the �bladder theory of corporate finance�� �The more the money
that builds up in the treasury, the greater is the pressure to piss it away.�

PE multiple of a stock

So, basically what Modigliani & Miller said was that if the company has got too much cash,

www.capitalideasonline.com
Page - 21
 Interview with Prof. Sanjay Bakshi

then management has no incentive to squander it away in useless or value destructive

projects. Well the reality is very different. So markets, when they sense that the company is
going to misallocate surplus cash on its balance sheet, punish the company by putting a very
low value to that cash.

I think dividend policy matters a great deal in the valuation of a firm. I am reminded of what
Mr. Graham said in the context of a PE multiple of a stock. What is PE multiple? The E is the
earnings. From that E, dividends are paid to the shareholders. You could think of the earnings
that are paid out as dividends as a stream of cash flow coming out which you can value
independently as a bond and that value will have a high multiplier. Why? Let see � today
AAA bond yield is about 8% and a bond which pays Rs 8 as interest should sell at par i.e. Rs
100 so the multiplier is 12.5, which is much higher than the multiplier at which most
earnings-based deep-value shares sell in relation to their earnings. The equivalent multiplier
on dividends will be higher because dividends are not taxed in the hands of the investors.

In firms which have a reputation of squandering away the cash in low-return projects, that
part of earnings which are paid out would be capitalized at a much higher PE multiple than
that part of earnings which are retained. And the weighted average multiplier will be equal to
the PE of the firm which the market is giving. Enter Mr. Munger�s concept of backward
thinking. You can take the market price of the shares of an over-capitalized firm stock having
a low dividend payout ratio. You can value the dividend stream separately, as if it were a
bond and reduce that value from the current stock price. What remains then the market�s
assessment of the value of the earnings retained � typically these will produce a much lower
multiplier than the multiplier on earnings paid out. So there are really two PE multiples in
such stocks. It�s elementary math. And then you can think about what will happen to this
company�s market value if the dividend-payout policy were to change (on its own, or due to
a little bit of nudging on your part!). I find that idea very useful indeed.

Lets look at VST Industries as an example to illustrate the idea of two PE multiples to a stock.
This company is valued i.e. its stock is selling at 330 rupees per share. It’s valued in the

www.capitalideasonline.com
Page - 22
 Interview with Prof. Sanjay Bakshi

market at 510 crores rupees. There is no debt and out of Rs 510 crores rupees, if you remove
the amount of cash on that balance sheet you find that they have got 160 crores in cash and
there is some other cash in the balance sheet which I’m not counting. I am looking at only
surplus cash. So you remove this Rs 160 crores from the market cap. But that Rs 160 crores
cash was in March 2006 and that figure would have increased by now. So if we look at the
quarterly results and just roughly speaking, can take the last four quarters of that we find
that another about Rs. 50 crores of cash has come in.

So if the company is valued at about Rs. 300 crores net of cash and on Rs. 330 market price
and they give you a dividend of Rs 12.50 per share. Now this Rs 12.50 of dividends is safe
given the nature of this business (cash generative, low capital intensity, low growth) we can
think of this as a perpetuity and if you can value this as a perpetuity coming in and it’s the
tax-free perpetuity, it is valued at probably about 5.2% p.a. tax free yield (AAA bond yield 8%
p.a. and tax rate of 35%). So we have a multiplier of 19.2 times on that part of the earnings,
which are being paid out as dividends. That comes to Rs 240 per share. Which means that
you are paying Rs 330 less Rs 240 i.e. only Rs 90 per share for that part of earnings which
are retained inside.

What would happen if you were to change the dividend payout ratio?

Now, what Graham showed was what could happen – or rather how I interpret what Graham
said what would happen if you were to change the dividend payout ratio? If VST were to
increase the dividend from Rs 12.50 per share at present, then an additional part of the
earnings which are now paid out will attract a higher multiplier and a lower part of earnings
which are retained will attract the lower PE multiple, so the PE multiple of the firm should
logically rise. Its elementary math� In effect, inside such stocks is a hidden bond component,
which can be easily valued, by us. The value of the hidden bond component will rise if its
coupons were to increase.

Now obviously that sounds very simple and arithmetical and elegant to say that in some
situations like VST if dividend payout ratio is increased, the stock�s PE multiple warrants an

www.capitalideasonline.com
Page - 23
 Interview with Prof. Sanjay Bakshi

increase as well. But an increase in the dividend payout ratio involves, from the perspective
of the management a reduction in corporate assets under their control and that’s not
something that managements typically like to do. MM assumed otherwise but they were
living in a make-believe world of perfect rationality.

So what works in theory and whether it will work in practice depends a lot on how persuasive
investors can become in getting managements to increase the dividend payout ratio in such
cases and I think there is a huge need for activism in that direction. I think institutional
investors in particular can play a role here, if Government of India can ask public sector units
to pay a specific dividend payout ratio because it needs the money, then why can�t it not be
the case with institutional investors asking (who control a large number of corporations
equity) to do the same thing? So for every one rupee of dividend jump, you will get
substantially higher jump in market valuation. So it’s not just the additional dividend that
comes in. You also get additional market value when dividends are hiked. You can have your
cake and eat it too!

There are other deep-value strategies arising out of dividend policy. Market prices are very
sensitive to dividend announcements and rightly so because insiders are giving out signals as
to what the future corporate earning power is likely to be. So it makes a great deal of sense
to always track first-time dividend payers or companies, which are restoring dividends after a
long gap. The concept of �sticky dividends� is important here. Stock prices are inherently
more volatile than underlying corporate earning power and earning power is more volatile
than the dividends which come from this earning power. Companies that announce dividends
for the first time, or hike dividends, are unlikely to cut them � they like to be consistent like
everyone else. Mr Munger�s model from commitment and consistency is applicable here.

It makes sense to look at companies, which are starting to pay dividends because if the
market price is very sensitive than halving a dividend from Rs 4 to Rs 2 will cause a
disproportionate decline in the market price. So it might become a bargain simply because
the company cut the dividend. So, from a value investor perspective, a company which is

www.capitalideasonline.com
Page - 24
 Interview with Prof. Sanjay Bakshi

doing badly for a temporary reason, might actually be good news because the cut in dividend
might give that investor a chance to buy the stock at a much lower price than he was looking
at earlier and conversely if the dividend is then restored there is a major upside there for
him.

So you get opportunities in stocks of companies that are cutting dividend, skipping dividend,
hiking dividend substantially or even announcing special dividends. There are some
opportunities which I have seen in capturing dividends, using futures market – initially when
futures trading came to India. There were some opportunities there because markets were
not that efficient and companies were announcing dividends and as we know the futures
prices are a function of spot prices plus an interest cost less the expected dividend before
expiry. So, if you can anticipate when the dividend is likely to come and if the dividend is
large in relation to the market price and you know which contract period it’s going to come
in, then you can effectively buy the spot and short the futures and capture the dividend. This
works if you are a little ahead of the futures market in adjusting futures prices in response to,
or in anticipation of significant dividend announcements. Sometimes this window of
opportunity is open for only a few hours, so it pays to be prepared and act when the time
comes.

Letting someone else take the dividend

Often we let someone else take the dividend. This is interesting. If you buy a stock for the
dividend, then you should be careful because when you are buying in the anticipation of the
dividend hike and a dividend hike occurs and the record date is still far away, and lots yield
chasers have jumped in or are likely to, then its very possible that the rise in the stocks price
will be several times the rise in the dividend per share. And once that rise happens and you
are in the stock you have to decide whether to hold on and take the dividend or to let
someone else take it by selling the stock just before it turns ex-dividend. Well, in my
experience, its often better to let someone else take it because if you decide to take it
yourself you are taking a big risk on the capital gains that are almost in your bag. And how

www.capitalideasonline.com
Page - 25
 Interview with Prof. Sanjay Bakshi

the hell does it matter what you call your return? It could be dividends or capital gains or
anything (as Shakespeare said �what�s in a name?�). You are investing for total returns
and while you focus on that there will be people who want tax free yields and divided
strippers and the like, and its often fruitful to let them do the stripping while you enjoy the
taxable capital gains.

Availability bias strategies

Now lets come to availability bias, which I think is fundamentally important. Mr. Munger said
famously, �The brain can�t use what it can�t remember or what it is blocked from
recognizing because it is heavily influenced by one or more psychological tendencies bearing
strongly on it. And so the mind overweighs what is easily available.�

I like to organize my thought process by revolving around various kinds of biases and one of
the key ones is availability bias. I think if you read the paper and you see all these news
coming out to you and it’s so available and it�s so recent and it’s so vivid and it looks so
important, and people have a tendency to react � rather over-react- to it. You see all those
news flashes on the screen there. There is some �big event� taking place. It looks very big
at that moment and one of my questions that I had asked before is that, if you pick up the
paper of some three years ago, and you pick up the headlines of that period and you see
what was looking so important then, and what really happened subsequently, you will figure
out what availability bias really is all about because things that are really temporary,
transient, they get exaggerated by the media and everyone else. They get hyped in the
media because it sells � that�s incentive superpower and incentive-caused bias. The media
people have their own interests involved. People overreact to things and when they overreact
to things, some people can money off those hyper over-reactive people. So as a theme itself,
it makes a great sense to me to look at news from an availability bias of other people’s
perspective.

I read somewhere once that people assess the frequency, the probability or the likely cause
of that event by the degree to which instances or occurrences of that event are readily

www.capitalideasonline.com
Page - 26
 Interview with Prof. Sanjay Bakshi

available in memory. That is very correct.

Mr Munger talks about vividness and recency. People over-react to something that�s very
recent and is extra-vivid. And crazy people make markets go crazy too. They go to extremes
on events, not just because of availability bias, but a big part of the reason for market
extremes is due to availability bias. There are many other reasons which we don�t have
time for to discuss but they are all combine and they produce these kinds of crazy situations
– Mr Munger calls them Lollapalooza events � I love that word by the way – and if you look at
these crashes that you had in May, you know, May 2004, May 2006 or even before that. At
that point of time it is hard to come to the conclusion that this was the time to buy. But it
was.

People don�t care about value, they care about prices

To everybody at that time it looks liked it was the end of the world and well, at the end, it
was the beginning of the world, so to speak.

I think the annual budget drama is also a good example. Every year we have this huge drama
that goes on and we have this huge analytical effort going into what this budget will do and
what it will not do and how important it is. I think there are fundamental flaws in the way
people interpret the budget and changes in government budgets.

First, how important is the budget? I don’t think it’s terribly important because if the
government has laid out the basic policy and there are no fundamental changes in direction,
it’s not going to change the value much. The value of a company is the present value of all its
future cash flows. But markets behave otherwise and they tend assume that any change is in
a policy is a permanent change, while the reality is that governments are also fickle minded
as investors.

No change is permanent. The only thing that’s permanent is change. So the government
keeps on changing its fiscal and economic policies and markets also believe and I think
government also believes that any change that they would make would automatically result

www.capitalideasonline.com
Page - 27
 Interview with Prof. Sanjay Bakshi

in, say, increased revenues because they assume that people will sit back and do nothing
about it. They don’t think of the consequences of consequences as Mr. Munger tells us to do.
So people organize their businesses in a manner that often defeats the very purpose of the
change in government policy.

Very quickly we will see what’s happening to dividend policy right now. Companies are
distributing dividend in March because they will escape the additional tax that the
government is talking about. So people miscalculate, Government’s miscalculate, companies
miscalculate, and investors miscalculate. It’s a huge drama going on out there and we would
like to watch it from a distance and we sometimes get opportunities there. But most of the
times, I think it’s a useless waste of time thinking too much about how important is the
budget and what’s going to do to corporate values or corporate prices? People don’t care
about value. They care more about prices.

So we have interesting long-short strategies arising out of availability bias because markets
go to extremes. We have opportunities in both directions. We have people who get over
excited about a certain concept or fad. It’s just the latest thing going on out there whether it
is ethanol or whether it is wind power or what have you – you have all these kind of new
things which keep on coming up and people go crazy and there is are also elements of social
proof, envy, deprival-super-reaction in all of this craziness. But I think availability bias
contributes vary substantially to the extremes that you see in market in both directions and
there is money to be made there.

My education in a share holding structure, which I used to totally ignore earlier, is fairly
recent and its been hammered into me by a friend who is a very smart investor. I referred to
this gentleman in my earlier talk also. He has taught me that you should look at the
shareholding patterns from various perspectives, changes in the shareholding patterns and
what’s going on inside the firm. It is very important to monitor what the insiders are doing
and what all is happening to the shareholding structure. It’s not independent of valuation. It
is independent of corporate valuation, but it is not independent of market valuation and we

www.capitalideasonline.com
Page - 28
 Interview with Prof. Sanjay Bakshi

are trying to make money in the market. So, I think if you have an overvalued – a grossly
overvalued stock, combine that with a large FII interest plus the availability of that stock in
futures and options plus insider selling plus weak accounting plus your anticipation of
slowdown in growth which is not factored into market prices, you have a very good
combination which gives you a potential candidate for shorting. Of course the shareholding
structure is important in holding companies. My initial comments on holding companies arise
primarily because they are boxes of cash controlled by a people, who are very well
entrenched and that�s a function of the shareholding structure. So that point need not be
repeated.

On the other hand if you have an ultra cheap company and you see insider buying and low FII
interest, you have a very good combination. So FII interest is important. Class of investors are
also important because the class of investors tend to behave in similar fashion and if they
take a macro bet then they want to leave the country because they get scared about some
plague of some other macro event which really has no relevance to corporate values, then
these stocks will get hammered. These stocks are what you call as �over-owned stocks�.
Nothing like that happens to �under-owned� stocks.

Obviously the idea of insider buying or the insider selling is asymmetric. They need not be
interpreted in the same manner because when an insider is buying, he usually buys because
he is bullish. But he may be selling for all sorts of non-value-related reasons. It�s not
necessary that he is selling because he is bearish. So, to that extent more weightage should
be given to insider buying than to insider selling. Of course blind copying of insider buying
can be very foolish because the insiders could be buying for the wrong reasons also. They
may be foolishly bullish. So you have to combine it with your own thinking, your own analysis
and if you come to the conclusion that it is a cheap stock and then you see there is heavy
insider buying, you have a confirmation coming into your original thesis. Whereas if you see
insider selling then it need not necessarily mean that you are onto something wrong. Modern
databases also allow us to track what other smart investors are doing. So if we find what we

www.capitalideasonline.com
Page - 29
 Interview with Prof. Sanjay Bakshi

think is a cheap stock and then when we look the company�s shareholding pattern and find
some people we respect, and if some more analysis shows that the investor we respect paid
a higher price than the current stock price � well that�s a very nice and comfortable kind of
place to be in!

And sometimes we find people whom we have never heard off in the shareholding pattern of
the companies we think are cheap, then what can we do with that information? Well, we can
search for what else such people own and then analyze those ideas. All my life I have learnt
the wisdom in �one thing leads to another.� So, having a process that allows one thing to
lead to another is very useful way of searching for ideas. Incidentally science works the same
way � much was what a scientist discovers happens accidentally while he was working on
something else! And many a great scientist had the conviction to drop what he was working
on the moment he observed something very strange and worthy of further investigation. So,
if great scientists can allow for one thing leading to another, why can�t investors do the
same thing? I think they can, and should.

Mean reversion strategies

I come back to mean reversion as I mentioned earlier. It’s a fundamentally important model
and the coin flipping experiment is obvious there. If you flip a coin 10 times, you might end
up with eight heads; you flip it 20 times you might end up with 15 heads. You flp it 100,000
times; you are likely to end up with about 50,000 heads, i.e. 50%. So, as you increase the
number of flips, the fraction which equals total number of heads/total flips sort of gets pulled
towards a number of 0.5, that’s kind of a magnet which is attracting and that’s the mean.

The tendency to revert to mean is very fundamental because cycles can be explained with
the mean reversion process. Of course there is a whole bit of psychology involved in these
cycles. People expand recklessly because they don’t think what the other guy is doing. They
are envious and they think that if they don’t expand, the other guy will expand and he is
going to lose (deprival super reaction), and they copy each other automatically (social proof),
and they get over-confident, and they all have big egos (excessive self-regard) � so you get

www.capitalideasonline.com
Page - 30
 Interview with Prof. Sanjay Bakshi

a mixture that produces gluts and the over expansion obviously means a decline in the price
of the commodity and that result in things going the other way.

The incentives to make are turned into incentives to buy. So you should buy out your
competitors instead of making because the Tobin�s Q (the market value of the asset/its
replacement cost) has become less than one. Tobin�s Q is very fundamentally important
model to have in mind for long-term investors because in at least cyclical stocks in
commodities where entry barriers are not there, why would Tobin�s Q remain three or four
for a long time? It shouldn’t. There will be more capital coming in that industry. And when
Tobin�s Q becomes less than one, there will be consolidation.

I think sugar will be sweet one day. It’s very bitter right now for an investor who bought in
some time ago. People have lost great deal of money in sugar recently, but valuations have
gone below replacement cost and they are making a commodity that people are going to
continue to use and everybody doesn’t use �sugar free.� I don’t think Indians are going to
switch to �sugar free� any time soon. So the product is going to be around and people are
going to eat it and these guys are politically fairly powerful and it looks horrible right now but
that’s what it’s supposed to look like if they have a low price. It is not supposed to look sweet.
So it looks bitter right now. But one day I predict sugar will be very sweet. It’s like
Templeton’s idea of maximum pessimism. How far are we from the point of maximum
pessimism? That’s the issue and the issue is not whether it’s a buy or not. The issue is how
far is the point of maximum pessimism and which companies have the staying power and
which ones don�t. The ones that don�t will go away because they can’t survive and that will
result in consolidation of an industry, which needs consolidation. So all those things are
important here for the mean to get reverted. Things will get attracted back to normal.

Stock prices don’t revert to mean, but stock returns do and this is the fundamental
arithmetical truth that Mr. Buffett mentioned when he said that bull markets and bear
markets can obscure mathematical laws but they cannot repeal them. If the underlying
return on capital is 15%, you cannot have returns on the stock market of 25% p.a. forever.

www.capitalideasonline.com
Page - 31
 Interview with Prof. Sanjay Bakshi

You will have periods of 25% returns and you will also have periods of 5% return, maybe
-10% returns or -20% returns. But in the long run the stock returns must mirror underlying
corporate returns and that return incidentally in India has been going up. If you notice in the
last several years, return on equity has been going up and it hasn’t been going up because of
pricing power, indeed we have less pricing power. It is going up because of efficiency gains
caused by competition. So unlike Peter Lynch who said that �competition is hazardous to
investors� wealth� Indian companies record post-liberalization shows they have fared much
better for investors because of more competition and not despite it. Its excessive competition
that is hazardous to investors� wealth and not competition per se, in my view.

The human mind also has another tendency. It tends to put an arrow at the end of a trend
line. Basically the human mind is not wired such as to think in terms of mean reversion. It�s
a natural tendency to believe that trends are destiny, but all trends are not destiny and it
takes a bit of deprogramming to internalize the mean reversion concept and to use that to
come to the conclusion that things will change. It looks bad now or this is too good to be true.
This is not normal and the normal is going to come. Of course, there are exceptions,
sometimes the trend line is destiny, and sometimes what you see as light at the end of the
tunnel is actually an oncoming train. There are industries out there, which are going to be
extinct and they should go extinct because somebody else came with a better mousetrap
and you see a trend line and a long-term trend of diminishing visibility. Look at what�s
happening in fixed line telephony for example. It is very negative sector according to Dalal
Street. But that doesn’t mean you can’t make money there. If you look at MTNL and if you
remove the cash that they have and look at the company�s market value right now you will
realize that the markets are extremely negative. If you work backwards and you net off the
cash from the market value of the company, you will find that this is value of the business
and how much the business has earned in the past and even if you project the declining
thing, basically the company doesn’t have to deliver much in the next two or three years to
make up all the money that you are investing if were you have to buy the whole company
and this totally ignores all the real estate that they say they have which I think they do. I

www.capitalideasonline.com
Page - 32
 Interview with Prof. Sanjay Bakshi

have seen a few of their buildings and they look very expensive, so the point being that even
though you are looking at the company and Graham said this, �There is no such thing as a
bad business. There are only bad investments.� So if the price is so low even a company
that is destined for oblivion, you can make money in such stocks. And you should not ignore
such stocks. Mr. Buffett would disagree with me I think on this point. We�ve been long-term
bearish on MTNL�s business but that has not prevented us from making money in the stock.

CIO: Time is the friend of the good business. Not the bad one. Buffett basically spoke of
making 20 investment decisions over your lifetime, because that would help you to focus on
the good businesses.

Raw material theme

Prof. Sanjay Bakshi: One theme, which we have come to recently, is what we call the raw
material theme. Why is that so important? Again my friend taught me this thing. He said that
if you look at the P&L account of the company, you find that they have these elements of
cost. They have basically the manufacturing cost, salaries and administrative expenses and
other overheads, depreciation, interest expense, and others. Out of these, the raw material
cost, which is part of manufacturing expenses, stands out for its significance in many
manufacturing expense as well as for its volatility.

If the raw material expense as a proportion of sales is significant as compared to the

operating profit margin, and if the raw material is a commodity and if it is a volatile
commodity then you’re going to find major volatility in the basic margin of the business
arising primarily because of the change in the price of the raw material and its significance in
the cost structure. So it makes sense to look at companies, which are hurting on the raw
material front for a very long time. Margins are going down, down and down because there is
a raw material, the prices of which have gone up, which you think is temporary and you have
reasons to believe that it is temporary but the market thinks otherwise because the market
does not believe in mean reversion. Rather the market has a tendency to believe that trends
are destiny.

www.capitalideasonline.com
Page - 33
 Interview with Prof. Sanjay Bakshi

So, there are situations where the current earnings of the company are hurting because
margins are going down and if things revert back to normal, if raw material prices were to go
down, then things will revert back to normal, you will have restoration of margin. In the
meantime the market value of this company also goes down because of the tendency of
markets to project trends to destiny and the assumption that this margin reduction is
permanent. I have a case here which came to my notice sometime back. If you look at this
company called Britannia Industries, this company in March 01, 02, 03, 04, 05 and 06, I have
got six years here, the EBITDA on sales for the last six years has been 11%, 21%, 14%, 15%,
15% and 12%. But look at what happened to the margins in every quarter from from June 05
onwards. Margin went from 17% to 16% then to 12% then to 9% then again 9%, and then 6%

Now part of the reason why this happened is because of competition from ITC. There is no
doubt about that. But a big part of the reason why it happened was because of wheat prices.
Now how likely is it that this is permanent and what if they went back to normal levels? Then
the next question that comes whether the jump in margin will be retained or will be passed
on because of the competition factor here. While I was analyzing this, I found another
interesting angle to the whole thing. I discovered Vinita Bali and I am fairly impressed by
what she is doing to the company. So from a business perspective, given the background
that she has and the kind of decisions that she is implementing, the dramatic rise in the
growth rate in the top line of the company changed after she came. What happened to the
company after she took over and what was happening before that? It’s a very major change I
think. So she knows, she recognizes that there is a problem on this front, but these are self-
corrective problems. These are problems, which have to do with regression to the mean and
there is nothing much she can do on that front. Yet she can do something on the competition
front and on the management front. In my view, she’s doing some very sensible things on
what really matters because the basic business is a high return on capital business and in
such businesses if you have all the capital that this company has, you should try to grow this
business. And you should put more capital behind and that’s exactly what she is doing. She is
doing joint ventures, she is introducing new products, doing all these little tiny experiments

www.capitalideasonline.com
Page - 34
 Interview with Prof. Sanjay Bakshi

in the product markets and some of them will work and some of them will fail. It doesn’t cost
much to introduce a new product, but if it fails, it goes out; if it succeeds, you have a winner.
So she is really slicing the market and I am not a marketing professor, but I know what she’s
doing is right here and it seems to be working. The top line of the company is growing, so you
have a very interesting combination of a stock which has fallen by 40%-50% from the peak to
Rs 1260 at present. If you have topline growth and you have restoration of margins, you are
going to see a phenomenal jump in profitability and when you see that happening, markets
again will put a trend and an arrow at the end of that line and you will see a major upside
there.

There are management disputes. These stories keep on coming that might be settled. But if
you look at the real issues involved, the disputes that they have are not preventing Ms Bali to
implement what really matters. So I am giving more weightage to what she is doing and on
what is actually happening. I am seeing the result rather than the disputes, which have a way
of getting settled. I am really optimistic that people will settle their dispute and even if they
didn’t, it won’t destroy the company.

Well, this company has staying power, it has debt free, has cash, good management and
professional management. It is doing all the right things.

Value +Momentum

One of the things we are working on now is to think about combining value and momentum.

CIO: How do you combine value with momentum? Do you bring in RSI or something?

Prof. Sanjay Bakshi: No. The essential idea is that there is a trade-off involved in buying a
cheap value stock and just sitting on it. We are buying it at the bottom or near the bottom
and there is a huge upside out there but you may have to wait for a long time to get that
upside. However, in that list of really deep value stocks that we have, if you observe that
there is some interest in that particular stock which would be reflected by the fact that there
are huge volumes, and there is a jump in the stock price.

www.capitalideasonline.com
Page - 35
 Interview with Prof. Sanjay Bakshi

Of course the price of the stock has gone up, but it hasn’t gone up to the point where it
ceases to be a bargain. It is still a very good bargain, but it’s not as good a bargain as it used
to be. That’s normally an indicator and that should be counted as an indicator to increase
your exposure as explained earlier in the context of frequency magnitude way of thinking
that I mentioned earlier.

Now everybody knows that it is a cheap situation and suddenly you notice that there is an
interest coming in. So the odds of return have gone up. The total return may have gone down
because prices have already moved up a bit. So when you take additional exposure, you are
going to earn a lesser return, but the period over which you are going to earn that lesser
return is likely to be compressed because of the momentum element involved here. So that’s
a trade-off. And I have seen that happen in some of our stocks. And one of the stock which
comes to mind was a Pioneer Embroideries which has gone from what we thought was value
to total glamour in a very short while. We went and met the company and we thought it was
an interesting play and we went and bought into the stock and it’s just gone up into glamour
land.

In this case, we saw the momentum and we ignored it. And when the momentum became
quite rapid, we actually exited and we left a lot of money on the table which is fine and not a
big problem because I am very happy to sell a stock which is out of my value range and go
back into something which is is much better value, than to continue holding it in expectation
that momentum will take it up and up and up because that’s very risky. So I have seen in a
number of occasions in the deep value stocks, the reason why they go up is because there is
momentum coming in and there is an institutional interest, there is money behind that stock
which lifts the price to a level which is higher than your original cost but still significantly
below what you think the company is worth. The way we are trained as value investors, we
don’t do any thing, we just hold. Well I now think that we need to change that mindset so as
to allow combinations of momentum strategies with value strategies.

CIO: How do you gauge momentum?

www.capitalideasonline.com
Page - 36
 Interview with Prof. Sanjay Bakshi

Prof. Sanjay Bakshi: What I’m saying is that we should take the essence of momentum and
put it in a value strategy. I think it’s testable and it should be done and my hunch is it will
work. In a total holding period, when did the maximum returns come? It came over a very
short period of time. And what was happening in the short period of time? Well, the volumes
were going up and the price was going up like anything. So most of the return is compressed
over a very short time frame, which is essentially a momentum kind of time frame and yes, in
most cases the price goes way beyond our sell price. So we ignore that part of the
momentum, which is in glamour land, but we try to be in that part of the momentum, which
is still in value land � and then FM comes in. Put more money behind it because the odds
have gone up.

Over-optimism

Last one is over-optimism. 80% of the drivers consider themselves to be about average,
including me. My wife doesn’t agree.

We all know what over-optimism is. We are basically, being analysts, supremely optimistic.
Why do sell-side analysts give more buy calls than sell calls? Of course, incentives are the
major reason including incentive-caused bias. But I think there is more to it. They are very
optimistic people and managements tend to under-weigh competitive threats.

Recently we interviewed a company that makes a global commodity with a big, dumb
competitor out there. As you know, Mr. Buffett says that in some businesses, �you can’t be a
lot smarter than your dumbest competitor.� Well, this was one such business, and, in my
view, that dumb competitor is China because many of the Chinese companies are managed
with the intention of gaining market share, and not necessarily to create wealth for owners.

So this company in China was on the verge of expanding capacity by a degree that would
have caused a global glut in the commodity and one of the things that we asked the
management was, what will happen if this new capacity comes to market? He said, well, that
Chinese fellow doesn’t know how to make his plant work and you know it’s very technical

www.capitalideasonline.com
Page - 37
 Interview with Prof. Sanjay Bakshi

thing and it’s going to take a long time to stabilize the plant and his plant is in jeopardy and
it’s not going to happen. So don’t worry about it. And of course the plant came on stream.
Yes it was delayed, but it started producing in large quantities and the commodity prices had
gone down and the company is hurting. But the question to ask, if we meet management
with supreme optimism, is why wouldn’t the competitor which has already sunk $500 million
in a plant spend a little more money to revive the situation that he wants to be in? He wants
to restore the efficiency of the plant, he wants to start production and there is problem and
there are engineers out there who will be happy to come in and fix the problems. It’s not a
non-solvable problem. So, it makes to ask the reverse question (Mr. Munger�s Backward
Thinking idea) instead of simply swallowing what an over-optimistic manager will feed you.

Security analysis

So, I think one of the main problems in security analysis is we tend to believe what the
management tells us and they normally tell us what they want us to believe. So it’s a nice
vicious circle. So we have to be careful on the over-optimism front. Investors tend to become
widely optimistic about certain sectors and we use backward thinking in a lot in those
situations. I think Mr. Buffett in 2000 gave a wonderful example. He said, take this company
(he was talking about Yahoo or something – a company with $400 billion valuation) and if I
want just a 10% return on it, how much cash it should disgorge next year to rationalize
today�s valuation and if it could not disgorge the required amount, then how much
additional amount it would have to disgorge in the year 2 or 3 or 4 or 5 0r 10 years to
rationalize today�s price?

So, he illustrated the powerful idea of backward thinking to figure out how much cash the
company needs to generate over the next few years, to justify today’s value. And from that
figure of required cash generation, you can work backwards further to find out as to what
kind of volumes the business needs to produce to generate the necessary revenues and what
kind of margins will it have to earn, to generate the required EBITDA, to have those kind of
free cash flows to justify today’s values.

www.capitalideasonline.com
Page - 38
 Interview with Prof. Sanjay Bakshi

CIO: You can reverse engineer the price and figure out the expectations, Michael Mauboussin
wrote about that in �Expectations Investing.�

Prof. Sanjay Bakshi: Absolutely.

Backward thinking really works because it helps you to be more objective as Mr. Munger
described. It helps you to see the lunacy in the market prices occasionally because it allows
you the privilege of reducing the problem to a discipline which is more fundamental than
your own discipline. By working that way if you can hit against the law of physics and come
to the conclusion and that there is no way a company can have the kind of numbers in year
10 which wil be required to rationalize today�s valuation because there isn�t enough
capacity possible to be able to produce the kind of revenues, to be able to produce the kind
of EBITDA, and free cash flows that will justify today’s valuations.

So when you come to that kind of a conclusion, you are really on to something and it works
both directions. It doesn�t just work on overvalued stocks. You may have even extreme
pessimism causing low prices and backward thinking in that sense really helps. So instead of
doing the elaborate exercise of projecting future free cash flow, which, I think, is just too
tough to get right, it’s way better to take the market value as Mr. Mauboussin mentioned in
his book as correct and work backwards to figure out what those assumptions are to make it
correct and then question those assumptions. Most of the times you will agree with those
assumptions which mean that there is the range of value and within that range you can
rationalize the value under normal assumptions. Within this range, you can very easily
rationalize the value. But when it goes out of that range, then it becomes harder to
rationalize and when it goes way beyond that range, it becomes just impossible to rationalize
it. So we have opportunities in that area.

Working capital bargains

CIO: Could you tell us something about ideas that you have discarded? You may have some
ideas which didn�t work and which you discarded.

www.capitalideasonline.com
Page - 39
 Interview with Prof. Sanjay Bakshi

Prof. Sanjay Bakshi: Working capital bargains. I have just discarded that.

CIO: Why?

Prof. Sanjay Bakshi: I tried it over and over again but it didn’t work out. I think Mr. Buffett also
mentions in his talks where he said that the working capital is not going to be there because
in the old days, in the US, if a stockholder found that the company is selling for less than
working capital, they could just take the working capital and go home. They could liquidate
the company for the working capital and take the money and go home. But you can’t do that
over here. You can’t do that in the US, you can’t do that over here because there are rules
and regulations which prevent a company from being liquidatied. So whenever you are
looking at a company from a liquidation standpoint, you should think carefully as to what are
the chances of that liquidation taking place. Value is one thing but the probability of realizing
that value is a totally different ballgame altogether. So if you think of the probability of
realizing that value is low, then the markets are going to factor that in the valuations. So in a
working capital situation, the company is working capital heavy, but the working capital is not
likely to be converted into cash and distributed to the stockholders, then it may remain a
bargain. So that’s not worked for me at all.

Holding Company

The other one that I mentioned to you was the holding company. I have become pretty
skeptical on that. I talked about that earlier. In the absence of a catalyst, investing in asset-
bargains is going to cost you hugely in terms of patience and opportunity cost. I know some
investors who have faced that problem.

So when you look at asset heavy companies selling at less than liquidation value then you
should look at, like perhaps my favorite would be Great Eastern Shipping. Look at the
company � they recently completed a spin-off and they have all these wonderful vessels and
the company is on its way to become debt free with the amount of cash that the business is
generating, it’s break up value is 400 bucks a share plus according to the management. And

www.capitalideasonline.com
Page - 40
 Interview with Prof. Sanjay Bakshi

the current stock price is Rs 192.

Well, I think the NAV is somewhere in the region of Rs. 350 a share and it’s easy to value
these pieces of assets because there is an active second hand market, you have to do is to
get a reasonably good hang of what the liquidation value is and that number is very large in
relation to the current market price of the stock and again the promoters are vulnerable
because I think they have only about 27 or 28%.

CIO: They are buying their own stock.

Prof. Sanjay Bakshi: But they are buying it in very small quantities. It�s very small. The point
is that on one hand they are expanding. And I think the fundamental question to ask is that
are you in the shipping business or are you in the business of making money for your
stockholders?

So if you are going to make money for your stockholders, then you have to deliver a return to
them on their investment and sometimes the best way to do that is to do a trade-off. You
want to expand, be a little more realistic in that and allocate a bit of the capital and your debt
capacity and your excess cash that is building up in your treasury to a very aggressive
buyback program and many of the best businesses that Mr. Buffett has owned have been
ones which have grown in size and profitability and at the same time have shrunken the
number of shares outstanding. Its analogous to owning a slice in a pizza which keeps on
growing and the number of slices keep on shrinking.

So I think that’s the trade-off, which managements have to acknowledge and understand and
I don’t think they do it. They don’t just get it unless it’s given to them in some forceful
manner which may happen or may not happen, who knows, but what matters is that there
are companies out there which have liquid assets, in the sense that they can be liquidated
very easily in the international markets and which are selling in the stock market at a fraction
of those valuations. And managements are not doing anything about it. So will this last? Well,
I don’t think it would. I don’t think it should also.

www.capitalideasonline.com
Page - 41
 Interview with Prof. Sanjay Bakshi

Diversification

CIO: As far as diversification is concerned, how diversified should you be?

Prof. Sanjay Bakshi: I have been thinking and pondering over this question for a long time
and I think I have the answer. At-least the answer which is right for me is that, as you should
be willing to figure out what are the odds of success are. It again boils down to frequency
magnitude. It’s such a fundamental concept. Graham bought cheap shares. He had no clue
what the companies were doing. He didn’t know whether the stock would go up or not. He
knew the chances of its going up was high but it was not a certainty; even a 60% chance was
good enough for him and he diversified enormously. So the probability of the portfolio doing
very well went up because they built up large number of stocks. On the other hand, Mr.
Buffett now in his current form wants certainties, which means if he finds those certainties
he�ll put a huge amount of capital them and if you find those certainties, then the amount of
effort that goes in identifying them and understanding them requires much more intensity.
So it depends on what you find. If you found something that is overwhelmingly likely to be
giving you phenomenal returns over a period of time, then you should back up the truck as
Mr. Munger recommends. So the diversification comes from your ability to identify
opportunities.

Now in my own experience in the past, if I look at my own track record, my personal track
record where I am not governed by the so-called diversification norms which my clients used
to ask me to adhere to, I have seen that the times when I have had those jumps in my net
worth, you know those spurts you have, they happen because of only a few transactions and
that happens only because I put a lot of money behind that particular idea. So, the Buffett
model of low diversification has clearly worked for me. But, by and large, I think Graham’s
model works enormously because it allows you the privilege of not doing too much intensive
analysis in a specific situation. You are doing statistical analysis. You are doing a kind of
analysis which doesn’t require you a great deal of insight into the products of the company or
the kind of competitive advantages they have, the kind of Porter analysis that is required to

www.capitalideasonline.com
Page - 42
 Interview with Prof. Sanjay Bakshi

be done in Buffett kind of a system. We don’t have that in Graham. So it all depends on
what’s right for you. In my case most of my best investments required me to put a bigger
amount of money, bigger amount of my net worth into a given idea, while at the same time
there have been ideas where I put a smaller amount of money. There has been a lot of
diversification and that has also served me well. But as Mr. Buffett, if he removes 10 of his
best ideas, the returns will become very sub normal and that is the case with me also.

CIO: Are you saying that you will put a large amount of your net worth into company which
may be cheap but are not �inevitable.�

Prof. Sanjay Bakshi: Those are inevitables because there is a catalyst and the catalyst
happens to be me or somebody known to me. So once you know that it is cheap, but there
are elements in place which will cause it to be not so cheap over a period of time, then the
condition comes from there because its you who are controlling the events. Because at the
end of the day the market value of a share will be only what somebody else will pay for it and
you have no way of predicting with any degree of certainty as to what that is or is likely to
be. But on the other hand if you are involved in a corporate event where you know that or if
you are talking to a company and you know advising them and telling them, look, the
company is very cheap and if you implement these three or four things, that might get your
company to be much better recognized in the market than is the case right now or you might
do things which will force the market to recognize that they are making a valuation error
which would have happen for example if you, as I mentioned to you convince companies to
dramatically hike their dividend payout ratios. In certain cases it will have an inevitable effect
of a sudden jump. When you are behind that kind of a situation and you are advising them
and you are hoping that they will agree and you are coming to the conclusion they are likely
to agree, then it makes sense to back it up with your capital.

CIO: Should one use leverage?

Leverage

www.capitalideasonline.com
Page - 43
 Interview with Prof. Sanjay Bakshi

Prof. Sanjay Bakshi: I would indeed use leverage. Graham had this wonderful definition of
what is an investment and what is speculation and he said that an investment operation is
one, which, upon thorough analysis, promises safety of principal plus an adequate return on
that principal. So you put the safety first and the return later, that�s obvious, but he never
talks about long-term versus short term. He never said that you should buy stocks on margin
or you could buy them with your own money. He never advised against about using borrowed
money and in an example he demonstrated that you couldn’t have made money unless you
had invested low-cost borrowed money in that deal investment operation. The situation was
such that it made sense for you to borrow money and do it. It was extremely low market risk
kind of a situation and the spread was so narrow, but the cost of debt capital was cheaper. So
it spiked up your equity return if you borrowed money and did it. Now if you go through many
such opportunities in life, occasionally you will find a chance to double or triple your own
capital in a situation if you put a lot of borrowed money behind it. So that’s what I have done.
But you better be sure about what you’re doing.

CIO: I believe Graham said, and I may be wrong, that an investor should have at least 25%
cash.

Prof. Sanjay Bakshi: He wrote that in the �Intelligent Investor� which was meant for the
layman. �Intelligent Investor� is not the book I would prescribe for people who are in this
profession. The book for them is �Security Analysis�. Even in the �Intelligent Investor� he
has in chapter on the enterprising investor who would take more risks because he could bring
more analysis on the table. So if you are going to be able to analyze situations more
thoroughly and if you’re able to identify opportunities, which give you a very good return,
then I think you should allocate more to equities. You should use leverage if the market risk is
not there. An event risk may be there. But even so when you use leverage, use it to the point
where even if things go wrong on the worst-case scenario that you don’t lose much. You
don’t have a solvency problem. You don�t lose much, which is exactly what Mr. Buffett does
in his insurance underwriting. You see if you think about it, the insurance underwriting

www.capitalideasonline.com
Page - 44
 Interview with Prof. Sanjay Bakshi

business is essentially borrowing. He says he doesn’t use much leverage, but I think most of
the money he gets from the float, is borrowed money. It’s just low, zero cost borrowed
money, which he has to eventually pay back. So he is using leverage. He is using other
people’s money and most fortunes of the world, if you really think about it, are made not by
your own money but the use of other people’s money and people use have creative ways of
doing that and Mr. Buffett has a wonderful creative mind and he has done that. It appears
that he has not used any leverage. But I think he has used a phenomenal amount of low-cost,
beautifully structured, long-term leverage and that zero cost leverage is coming from his
genius of having people like Ajit Jain working with him and giving him a zero cost float to
work with. Now if you are really conservative, you could take a zero cost float and put it in a
treasury bond and capture the difference. But he puts it in stocks. He puts it in operating
businesses. And he makes 30% against the zero cost. So you know that shows up in his
equity.

Risk & Uncertainty

CIO: What is the difference between risk and uncertainty?

Prof. Sanjay Bakshi: Basically Mr. Buffett lays it down very elegantly in his 1993 letter. He
wrote down three or four things on the certainty in which you can estimate the future
prospects of the business, the operating skills as well as the capital allocation skills of the
management, the integrity of the management, and the inflation and the taxation that would
reduce the gross return.

Those are the four or five things he listed and as he said, these are not things that you can
measure statistically or download from a database. There is no database that will throw a
number at you to measure its riskiness. He also showed that the same investment could have
different risks profiles for different investors. If I don’t know enough, then it may be more
risky for me than it is for you.

So you can’t use a number called beta or some other number that tells you that this is the

www.capitalideasonline.com
Page - 45
 Interview with Prof. Sanjay Bakshi

risk of this investment because the risk is different for different people.

The other thing that Mr. Buffett said in his talk at Columbia (Superinvestors of Graham-and-
Doddsville) that in value investing, the less the risk you take the more the return you can
expect to make. And that obvious common sense but counter-intuitive because most people
think that you can�t increase returns without increasing risk. Well, that�s not the way it
works. As Mr. Buffett said, to increase returns you must reduce risk � risk as he defined and
not as modern academic finance defines it.

So you are trying to control risk. There is no doubt about that. Now, what is important in that
is that how sure are you that you are buying something cheap. Graham never tried to value a
stock. He only tried to ensure that he had significant margin of safety, which is a far easier
job to do than to value a stock. So he just wanted assurance that the price that he was
paying was much lower than the value he was obtaining through his purchase. He insisted on
a large margin of safety without specifying how large it should be. And, after a point it stops
mattering howlarge it should be. So he was perfectly happy to sell stocks below fair value
because he found something else that was even cheaper. So he won’t wait for the actual
value to come because he doesn’t really didn�t know what the value is. So from the risk
perspective what matters most is that you are buying cheap things which means that (1) you
have odds in your favor; and (2) you are not putting too much money behind a given idea so
that if it turns out to be wrong, you don�t suffer grievous harm.

Global events or some systemic events or the business getting destroyed because somebody
invents a better mousetrap, something which you even cannot anticipate at that time of
investing�those risks you have to assume away through diversification. So you have this
two-pronged strategy- buy cheap and diversify and if you do these two things, then it’s OK.

But you can also get into the mindset where you don’t want to diversify, you want to focus,
you want to put a lot of money behind of a given idea then you have to do Mr. Buffett�s way
of analysis where you have to really look at the management, you have to look at the product
the prospects. You have to look at the operating skills of the management and that is not

www.capitalideasonline.com
Page - 46
 Interview with Prof. Sanjay Bakshi

enough because there are a whole lot of companies out there with phenomenal operating
skills but are zero in capital allocation. They don’t know what market cap means. They don’t
know what a buyback does. They don’t know that buyback can compete beautifully against a
capital-intensive project and deliver returns to stockholders far in excess of what a wonderful
capital-intensive project would. So, it’s better not to do that but just to shrink the company
and buy back the stock. So they don’t know all those things and they don’t know dividend
policies. They don’t know how you can hike the dividend payout ratio and have a much better
return to the stockholders. So these capital allocation skills are not very well known. So it’s
not that people have bad intentions. It’s just that they don’t know that these tools are
available and they don’t know what really matters. It doesn’t matter that you don�t really
want to have the biggest company, or the biggest market share. You want to produce the
maximum amount of money, that’s what you’re there for. So I think there is some hope there
because if you are able to convince at least some management that look at these things. I
think we should try. As investors isn’t it our duty to approach management as capitalists and
say look, it’s my company and this is a way you should be doing things and be very
persistent about it?

We have role models like Carl Icahn and others. How do they do it? They are so persistent. So
I think you have to be persistent, you have to have capital and you have to just keep on
going after it. It’s not just the operations, it’s the allocation of capital decisions where
corporate value maximization potential lies its not that they don’t want to maximize value
but that they mis-associate value maximization with sales maximization, with market share
maximization or customer satisfaction maximization which are the wrong objectives. The
might be a means to an end but not an end in itself. You shouldn’t confuse the means and
ends. The end should be to create value. You know how, in the last speech in the wonderful
movie, �Other people�s money�, �Larry the Liquidator� demolished the argument and
explained what capitalism really is about. Everyone should see that movie and that speech.

CIO: I am very impressed with Edward Lampert because he is very young. He�s just 42. I

www.capitalideasonline.com
Page - 47
 Interview with Prof. Sanjay Bakshi

think he is already worth $1.8 billion. Now if you want to get rich in the stock market, you
have to be an activist.

Prof. Sanjay Bakshi: Have you read Martin Fridson, �How to be a billionaire?� There are
many chapters in it each one organized around a theme. He basically went to all the
billionaires and he found out what did they just do. I mean converted them into different
chapters and put them in a nice book. One of them was on activism.

Being an activist it doesn’t mean that you take on the management. You could be proactive,
you can make suggestions, you can be persuasive, you can approach the management and
sit them down and let them understand as to where they are going wrong from the
perspective of an educated investor who can see things on a broader canvas than what they
can see. They have to realize that they are not in the textile business, they are not in the
steel business, they are not in the airline business. They are in the allocation of capital
business and that’s the job of a portfolio manager and that if you approach the management
in a humble manner and with an educated idea of what that company is really about, I don’t
think it’s impossible.

Opportunity cost

You know opportunity cost, you can�t think of it the way academics wants you to think.
Because they think that you should know the expected returns of every security out there
and that’s how you estimate opportunity cost. But that’s not the way. So when you think of
corporate capitalization and opportunity cost, it’s a fundamental economic concept, you have
to bring in certain level of confidence. You can’t do without it. You have to bring in satisficing.
So you know that these are the areas within which I have a competency. So within this area if
I do something then the cost of my doing something is also the next best thing. I could have
done within that. So there are lots of things, which are outside that universe which I don�t
know and I didn’t buy into like that Unitech IPO. Does that make me envious of that situation?
I don’t think so. It was outside my understanding so I didn’t do it. But it happens sometimes
that you make mistakes of omission on a scale that is grand in an area where you have a

www.capitalideasonline.com
Page - 48
 Interview with Prof. Sanjay Bakshi

competence. You just missed it, you just it whiz past you. Now that’s opportunity cost � the
cost of a missed opportunity that was available to you.

There was a book, �Contrarian Investment Strategies: The Next Generation� by David
Dreman. He has a chapter in it in which he says that you should buy the cheapest company
in the hottest sector which is a bit similar to the momentum thing which I mentioned in
�Value plus momentum strategies�. He liked the idea. I have never done it but I’d love to
experiment and see whether it works over here or not. Because in hindsight it is easy to say
that real estate will become hot.

However, if you are a deep value investor, you are usually trained to not look at things which
have gone up a lot already. Some people can train themselves out of it like Peter Lynch did.
He talked about the foolishness of ignoring a stock simply because its gone up.

A stock that has doubled or trebled does not appear cheap anymore. Well, that�s anchoring
bias � one is latching on to the wrong anchor of past price. The correct anchor should be
potential value, and not some past price from which it has risen. And reverse is also true.
Simply because as stock has fallen 80% from is past peak level does not necessarily make it
cheap. It may have fallen 80%. It can fall another 100%. There are many dot-com speculators
who averaged down their portfolios all the way to zero.

CIO: You spoke about your interest in psychology and what are the important mental models
you have taken from there. Mr. Munger did mention a few in his book.

Prof. Sanjay Bakshi: Six of them are from Robert Cialdini�s book, �Influence: Scence and
Practise� – Reciprocation, Commitment and Consistency, Social Proof, Authority, Liking,
Scarcity and then Mr. Munger expanded that list vastly in his marvelous talk on psychology.
Obviously the combinatorial effect of several of them is enormously useful way to analyze a
lot of things. So is the idea of observing something extreme and then working backwards to
identify the models which combined to produce that extreme outcome. It gives you a huge
insight and predictive abilities about what could happen in the future. The mental model

www.capitalideasonline.com
Page - 49
 Interview with Prof. Sanjay Bakshi

approach is not just limited to psychology. I think it should be expanded to other disciplines
as Munger recommends.

Biology, Business innovation & Evolution

CIO: Which would be the other disciplines that you are using in your mental models?

Prof. Sanjay Bakshi: Biology would surely be one. The essence of evolutionary biology is very
much applicable in understanding lots of things in the business world. At the end of the day
the businesses exist to survive and prosper and propagate just like species do and there are
new industries that will come up.

The model of fitness landscape from evolutionary biology is hugely important to understand
because the landscapes change. You know landscapes of various industries change all the
time and the organisms in the industry – species within that industry which could be those
companies that are most adaptive to the changes in the environment are the ones that are
going to survive. And the fitness landscape model is very essential to understand because it
tells you that there is a point beyond which you can’t change any more because you�ve
reached the top of your peak and the next peak you have to come down. So in some
industries you find that the way the business is run is so different from the way your business
is being run. There is no way you can compete with the new business model and you are too
well entrenched in your own way of doing things that you are eventually going to become a
niche player. You cannot remain competitive.

Also, I feel that you cannot understand innovation properly unless you understand Darwin�s
ideas on adaptation. It�s the same thing � the same ideas that come from evolutionary
biology that can help us understand innovation in the business world. Evolution is about
survival of the fittest and how you fit into your environment when environment changes. So
it�s just that sometimes you happen to be in the right place at the right time and you just
got lucky and then things worked. So if you’re innovating, you do this little experiment which
is what evolution does. You know little changes, copying errors from one generation to the

www.capitalideasonline.com
Page - 50
 Interview with Prof. Sanjay Bakshi

next one. Some of those copying errors which are random in nature and there is an element
of randomness from one to the next generation. Some of all those copying errors would
become improvements and then the randomness stops because from then on, if it is an
improvement it will propagate and if it propagates then it will become prosperous and that
species is going to become more and more prosperous and that applies to industry on a
grand scale. That�s exactly how things work in the business world too.

CIO: There is a very good book I don’t think if you�ve have read it �The Origin of Wealth�.
It�s on the new economics. The author has taken models from biology rather than physics to
explain economics.

Prof. Sanjay Bakshi: No I have not but now I will.

CIO: Many thanks.

www.capitalideasonline.com
Page - 51