Stray Thoughts

How do the managers try to predict stock prices?

2011-09-12T12:36:00.000-07:00

Why do managers want stock prices to rise? A flippant but, perhaps, not entirely incorrect answer would be so that they, the holders of stock options, can get richer. But the managers are given stock options to encourage them to take such decisions that would increase the shareholders’ wealth – the raison d’être for a firm.

High stock price does other good things too. It inspires the confidence of employees, suppliers and customers; it enables a firm to make strategic acquisitions, and on a more mundane level it allows a firm to raise both debt and equities at lower costs, reducing a firm’s cost of capital.

It seems trying to predict the stock price movement is just as important for the firm’s managers as it would be for its investors. Typically the managers try to bolster stock prices by investing in positive NPV projects, based on a cost of capital usually derived from CAPM or some other equally imprecise asset pricing model. But unfortunately the investors work out their own estimates of NPVs of a firm’s projects and these can vary widely from the managers’ estimates. Managers try to bring convergence in these estimates by arranging analysts’ meets and explaining the firm’s operations.

Analysts’ meets work both ways. Managers explain their actions and the likely impacts of those actions on value of the firm; managers also try to understand how the analysts think those actions will impact the value. Analysts, at least some of them, represent large investors – real or potential. Impact of large investors’ operations on unrelated (i.e. unrelated to the market wide price movements) fluctuations in stock prices is well documented. It explains why managers may hold analysts’ meets before major announcements.

Better still, managers may even try to understand the behaviour of their large investors just as much as they try to understand the preferences and predilections of their major customers.

Pre-money valuation and post-money valuation

2011-07-05T06:34:00.000-07:00

Some time back I had received a mail asking about pre-money valuation and post money valuation. Today teh corresponednt mailed me again saying that pre-money valuation should be equal to post money valuation less cash infusion. I know this position is taken by many professionals. But consider.

Suppose a firm today has value V and after it receives cash C its value becomes V1. I am simply saying that V1- V should be more than C. (If we say that pre-money valuation equals post money valuation less cash infusion then V1-V will equal C.) If we believe that firms make positive NPV investments it should be obvious. If the firm (at value V) gets cash C and makes positive NPV investments the incremental investments so made will add a value to its existing value V which would be more than C (because investment is positive NPV). It means C < V1-V or V1-V >C

Pre-money Valuation will equal Post-money valuation less cash infusion only when additional investments bring zero NPV; and in that case why should the firm seek to raise money?

PPP Transactions in Housing Development in India

2011-06-26T06:15:00.000-07:00

In all Indian cities vast chunks of lands are owned by governments, variously called Revenue Land, Municipal Land, Khas Mahal etc. Many cities also have vast stretches of land belonging to statutory entities such as railways, airports, port trusts etc. Rising employment opportunities in cities over last five decades has spurred rural to urban migration and as urban planning had so far overlooked the housing needs of lower income groups the migrants built shanties on the lands belonging to government / statutory entities, giving rise to slums. Cities were unable to evict the illegal squatters for various reasons. The most vociferously cited reason is vote-bank politics, but it is now generally agreed that the fundamental forces supporting continuance of slums were / are the cities’ dependence on the migrant labour living in the slums and their (cities’) inability to provide the slum dwellers decent housing elsewhere.

Because of their insecurity regarding tenure the slum dwellers built shanties with low cost material and to the extent possible with removable parts. This precluded vertical construction. Once the governments agree to letting go of their land, redeveloping the slums in most cities, where land prices are high, becomes financially viable if existing population were to be moved into proper dwelling units constructed in G+3 (or higher) apartment blocks, freeing most of the land presently occupied by the shanties. Part of the land so freed can be used for development of basic facilities - parks, drains, sewerage, roads etc. and the remaining parts can be sold to the real estate developers for commercial use. In an innovative step the two aspects – development of decent dwelling units for slum dwellers with all urban facilities and development of commercial space was clubber together by asking the real estate developers to cover both aspects. The tenders floated for this purpose often asked the number of free dwelling units that a bidder was to provide. Its arithmetic relates to cost of construction, cost of land at the site, area of the land being freed for commercial use and cost of commercial / HIG real estate in the neighbourhood. In some cases it may be possible for bidders to offer to accommodate all existing slum dwellers and pay some premium to the city government. In most cases a bidder would ask for some funding and the contract would get awarded on the basis of the lowest funding sought.

This redevelopment model, in which both public sector and private sector participate and the public sector hopes to benefit from the better project management skills of the private sector, has been called innovative PPP structures; innovative because normally PPP structures entail the private sector to run and maintain the assets created for their economic lives or for such long periods as would enable the private sector to recover its capital costs from user charges.

A PPP project, in road, port, power etc., is based on an agreement between a Government or statutory entity and a private sector company “for delivering a service on payment of user charges. The private sector partner is required to finance the project, and recover the capital costs over time, through user charges / annuities. Typical concession period ranges from 1 5 to 30 years”, even longer at times. A contract where the contractor receives payment for the capital cost at an earlier stage (such as on successful commissioning) and not based on life-cycle costs that include maintenance for the period, is not eligible, to be considered as a PPP transaction in terms of MoEA guidelines that are framed for developing basic infrastructure, such as roads, ports, electricity etc. in partnership with private sector.

In housing sector, however, the private sector partner, usually, seeks to recover full capital investment on completion of construction; and such projects would not be considered as PPP projects if the life time cost criteria were to be applied. To be considered a PPP project, in accordance with the aforesaid definition of PPP, the developer (private sector partner) should own the asset over its life time (say 50 years), maintain the houses over this period and recover the capital costs by way of user charge i.e. the rental. The real estate developers usually would not go for such transactions.

With further development of financial services sector, it is not unlikely that some real estate funds, in future, may look at owning and operating the housing assets. If such funds were around, the public sector could have co-opted them as private sector partners for developing housing projects and the project company (owned jointly by the real estate fund and the public sector entity) could get the assets created by farming out the construction work to developers / contractors; this is essentially replicating the PPP model for, say, power generation facilities – the project company runs the power plant but construction is carried out by contractors engaged by the project company for this purpose.

Till such real estate funds appear and true PPP models (involving life time costs) in housing sector can be developed, to encourage private sector investment in housing “innovative PPP models” for housing sector are increasingly employed by city governments across different states in India, notably Mumbai.

Rent Control vs Real Estate Regulation

2011-06-02T22:14:00.000-07:00

The verbs “to control” and “to regulate” have very similar meanings. So much so that many a time dictionaries define “regulate” as to control. Still control (and thereby controller) is considered bad and highly avoidable; and regulator is considered good, even desirable. It seems important to make an attempt at articulating the differences clearly.

Control usually implies a dominating influence, on the other hand “to regulate” implies to direct according to rules / principles / laws. When an executive entity makes rules (under laws made by legislature) and also administers those rules, the entity enjoys a dominating influence. On the other hand, if an executive entity makes rules and another independent entity administers them i.e. issues directions and decides on infractions of those rules, the second entity is seen as a regulator rather than a controller having dominating influence. But this distinction does not apply universally. Many a time the regulators issue regulations / guidelines and administer them; two examples are security market watchdog SEBI or the central bank RBI. It would appear that the difference is in degree rather than kind.

It is almost unanimously believed that for individual initiatives to flourish the state needs to create an environment where individuals are largely left alone. However, no one advocates for complete laissez faire. It is believed that a regulator, independent of the government, is perhaps the best (or the least bad) solution. Control is bad because by its nature it is intrusive and kills initiative.

Rent control in India, as in rest of the world, was introduced during the war years. The original objective was controlling inflation, which tends to go wild during wars. With rising prices landlords demand higher rents, tenants demand higher wages and so on. All rent control measures were introduced for a limited period, with words like “will cease to have effect after *****”. Politicians perhaps found it a very useful measure, after all tenants generally out-number land lords. Over the years the Rent Control Acts of some states were amended to exclude houses beyond a certain rent level from its purview. Some states have done away with such Acts altogether.

Rent control acts generally fix the rent by some formula to arrive at a rent (much) lower than the prevailing market rent. They generally disable the landlords from seeking eviction of tenants except on some specified grounds and they allow meagrely increases in rentals with passage of time.

Intuitively it would appear that people who could, would not invest in rental housing stock and that those owning spare houses would hesitate in offering those houses in the rental market. Researches have confirmed the second proposition. About the first, the findings are mixed. (Some researchers have observed that rent control measures have not affected the creation of rental housing stock in some cities.) The High Level Task Force on Affordable Housing for All (Task Force) appointed by the Ministry of Housing and Urban Poverty Alleviation in 2008 has identified in its report “modification in rent control act to make it compatible to owner and tenant and fiscal incentives for investment” as one of the required policy shifts that would encourage private sector to engage itself in production of rental housing stocks. A common theme in all States’ housing and habitat policies (that I have seen) is “to make appropriate amendments in Rent Control Act”.

Many arguments have been made against rent control. “Rent Control Laws in India – a Critical Analysis”, a working paper published by National Institute of Urban Affairs in 2006, written by Satvik Dev (coordinator: Paramita Dutta Dey) contains a well-structured, lucid summary of arguments against rent control. Dev has divided the arguments in three groups: economic, social and legal.
Among the economic arguments he counts: a) disincentive to those wanting to invest in rental housing, b) deterioration of existing housing stock, c) withdrawal of houses from rental market leading to reduced supply, d) reduced liquidity in the ownership housing market as tenanted premises are difficult to sell, e) erosion of municipal revenue as municipal taxes are largely based on rental value of houses, f) emergence of a large black economy from illegal renting

“The social impacts of Rent Control Acts” writes Dev “are more explicit and often, very bizarre. In absence of any fresh supply of rental housing, the existing tenants sit tight and the new entrants are the worst affected. The only possible way for them to get an apartment is through the death of an existing tenant (assuming no inheritance rights)” and thus young couples are forced to with their in-laws.

Among the legal arguments against rent control, he enumerates: a) Landlord’s right to seek eviction (available in Transfer of Properties Act) gets diluted, b) the provisions of Rent Control Acts have often been ineffective as significantly large groups have been excluded from its purview, c) Apex Court has many a time denounced some provisions of these acts as unreasonable, d) most rent control acts have been poorly drafted leading to problems in implementation.

Instead of having rent controllers I feel we should have rent-regulators or even real estate regulators. And the prime function of the regulator should be proper administration of the tenancy agreements entered into by the tenants and landlords. Except during some emergent situations (war for example, or an economic emergency brought about by very high inflation rate) State should not re-interpret the terms of a tenancy agreement and must not assume that any party needs its sympathy. If the contracting parties were competent to enter into it, the contract’s sanctity must be upheld. (I find it particularly odious that a government increases dearness allowances to its employees with rise in price index but restricts the right of a landlord to demand higher rents in view of rising prices.)

Instead of rent controllers we need real estate regulators. The Task Force in its report has favoured setting up of a "real estate regulator", on a priority basis, that could serve as a single window for overseeing and monitoring the affordable housing agenda, regulate the activities, promote policy reform e.g., stamp duty and registration, protect the consumers from real estate fraud and coordinate digitisation of land records. (I have not been able to learn the Ministry’s reaction to this recommendation of the Task Force.) I feel the regulators’ role should be more consequential than merely facilitating private sector’s entry into the low value segment of the real estate sector.

I believe houses in rental market need to be seen as part of urban infrastructure and need to be regulated like any other infrastructure facility such as toll road concession or electric supply utility. In all three cases the owners (of houses or the concession or the power supply rights) enjoy varying degrees of monopoly and unless regulated would demand monopoly rent, distort the market and lead to sub-optimum allocation of society’s resources. On the other hand ham handed attempts at controlling their monopolistic tendencies have proved to be equally harmful. We have seen it in the past – in low power generation capacity and in slow increase in housing stocks against burgeoning demands.

Particularly in housing market the checks and balances introduced by the governments have often been ad hoc in nature, as if the problems were transient and self-limiting that would go away with time. We need permanent, independent real estate regulators in each State, with judicial powers; and against whose orders appeals cannot be made in a court below that of the District Judge. I feel eviction of a recalcitrant tenant should be the chief concern of the real estate regulator and if this were achieved we should see a whooping increase in rental houses, built not by big time real estate firms but by our friendly, neighborhood landlord, who at present is investing in stocks.

Can we make the real estate market “efficient”?

2011-05-30T00:00:00.000-07:00

Often, a market is called efficient if the prices reflect all available information. Since buying and selling are serious businesses in every deal both parties do consider all information they possess before asking or offering a price. This implies that in all deals the prices reflect all information available to the agents; and in a market free of controls each deal is struck at the price that the agents, based on the information they possess, consider most advantageous to them. And as a corollary, if we go by the first line of this paragraph, all markets are efficient.
Clearly we need to build more on the axiom that in efficient markets prices reflect all available information. We add a rider, in efficient markets relevant information i.e. the information that may have a bearing on prices, are shared by many if not most of the participants. If a (price) sensitive information for an asset is known to only a few people, those few people will be able to price that asset correctly and benefit from its mispricing, if any, in the market. If that information were widely available the price of the asset will quickly adjust to the correct level i.e. the market will not be mispricing it. This makes wide availability of information an essential precondition for an efficient market.
But wide availability of information, though necessary, is not sufficient. There should be many participants willing to trade (i.e. buy or sell) using that information. Willingness to trade (and thus make profit) diminishes as the cost of transaction increases. Willingness to trade also diminishes if the transaction cannot be consummated soon. We do not like to make commitments and then wait.
Thus, three features that contribute to making a market efficient are: wide availability of information, low transaction cost and fast conclusion of transactions.
Real estate markets in India lack all three features. We generally have little idea of demand and considerably less of supply positions; with lawyers’ fees, registration fee, transfer (mutation) fee etc. buying and selling real estates are very expensive and the process is tortuous and time consuming.
Compare it with stock market or commodities market or even wholesale grain markets. Efficient markets reduce the cost to the end-user. It does so by forcing the intermediaries to accept a lower profit on each deal. In stock markets the broker’s charges are usually around than 0.5%. In rental housing market, in Mumbai, the prospective tenant pays two months’ rental as brokerage on an eleven months “leave license agreement”; that amounts to 18% brokerage, increasing the end-user’s cost by that much. It is even worse in land market, where we do not come across regular brokers but “investors” who buy chunks of land and sell later. These market prices are reflected in high value of urban land, so much so that economically weaker sections, low income groups and parts of middle income groups have been squeezed out of the urban land market.
The High Level Task Force on Affordable Housing for All (constituted by the Ministry of Housing and Urban Poverty Alleviation in 2008) observes in its report “Urban land markets do not operate efficiently – land is in short supply or the price of land is too high. In fact, the land market, in the sense the term is generally understood, i.e. a place where land can be purchased and sold, does not exist in India. A large number of transactions, therefore, take place under non-market conditions, which work against urban poor households.”
The Task Force seem to suggest that many transactions would take place under “market condition” if there were a “land market, in the sense the term is generally understood, i.e. a place where land can be purchased and sold”. I am not sure that specifying a particular place for such transactions would help the matter in any way – in fact it may even exacerbate it. Those with memberships to such a “place” will demand monopoly (oligopoly) rents. In fact we must note the differences between “market’ and “market place”.
To improve the land market’s efficiency, I think, we should initiate steps that would take us nearer to the three pre-conditions of an efficient market: wide availability of information, low transaction cost and faster and convenient completion of a transaction. There are evidences that the transaction costs are coming down at the lower end of the market. Some States have drastically reduced the registration fee for “affordable housing”. But in all such cases there are severe restrictions on reselling. Don’t take me wrong, I think the restrictions on reselling are perfectly justified given the subsidies (other than low registration costs) that bring down the cost of these houses within “affordable” limits. But nevertheless, efficiency of the market does get affected adversely dues to these restrictions.
While reducing cost of transaction will have revenue impacts for the States, steps to make more information more widely available do not entail any such impacts and, perhaps, can be initiated sooner. About availability of information the Task Force has observed that “the piecemeal information presently available in - most cities/wards is derived from the issue of commencement and completion certificates. This crucial information needs to be collected and collated through an institutional structure”. Perhaps, we can have a central registry from where such data (number of houses being constructed with some details as to their sizes, locations and other standard features) can be accessed free of cost by everyone. The present state of computerisation in the local governments can easily support creation of such a central database.
This however does not resolve the problem of opacity in availability of land. Residential plots for individual households in most cities are developed and sold by State Housing Boards. Housing Boards suffer from the procedural rigmarole typical of public sector. Even if it is announced that a Housing Board is developing a large tract of land in a city, no one can really guess when the plots on that land will actually be entering the market. It can be examined whether the task of developing residential plots can be undertaken in Public Private Partnership mode, where the private sector undertakes the development and plotting work in accordance with the cities’ CDPs and gets paid for it by the Housing Boards. Superior project management practices of the private sector will make it easier to estimate the supply of lands in future.
Further, all information on sales and purchases of real estates – land or houses - should be made available on the central registry on daily basis.
If information were made more widely available and transaction time were reduced, I think, the number of transaction would increase that may enable the States to consider reducing the registration charges, change of land use charge etc. that constitute sizeable chunks of transaction cost.

Rental Value and Required Return from Investment in Housing Stock

2011-05-13T12:43:00.000-07:00

What kind of return would a landlord expect (require) on his investment in housing stock? Let us take two commonly considered returns available in the marketplace to an investor. Government of India offers around 7.5 per cent on long term debts; and it entails no uncertainty. Stock market has generally returned over 16 per cent but with considerable volatility. Stock market’s return includes capital appreciation of stocks and its volatility makes the returns uncertain. Let us assume that considering the relatively less uncertainty of real estate markets (compared to stock markets) investors would require around 15 per cent return and that these days about 12 per cent of it would come from capital appreciation. (The value of land generally goes up but the value of structure falls over the years and capital return, of 12% as assumed, is net of reinvestments required to keep the property in good repairs.) Now do not lose sight of the fact that about 30 per cent of rental would go towards municipal taxes and that income from house property is taxable unlike income from stock markets – either dividend or capital appreciation if held for more than two years. The numbers would start falling in place:

It seems landlords would require rental to be about 6.5% of the capital value of their property. I am living in a rental flat in Bhubaneswar, rent Rs 14,000 per month. It works out to Rs 168,000 per annum and 6.5% required return would indicate capital value around Rs 2.6 million. I am assured flats in this development are selling at Rs 5.5 million. That suggests a low required return of 3% and got me thinking again. It seems municipal taxes in Bhubaneswar, though 30% of rental value, are determined on the basis of some standard / fair rent computed by municipal authorities, which is substantially less than the prevailing rental.

On PE Ratio

2011-05-10T04:21:00.000-07:00

Some time back I had come across this question: “Meaning of P/e ratio and its significance? If the co is having a P/e ratio of 20,then what does it denote and also 20is expressed in what like yrs, time, %, etc?” I share here the reply I had given.

A few decades back milk yielding cows were traded at Rs 1000 per litre. (I do not know the current rate but I am sure it would be higher). What this rate meant that value of a cow that gives 1 litre of milk daily would be1000 and of another that gives 12 litres daily would be 12000.

Now think of an equity share as an income generating asset just like a cow. Whereas the cow gave milk the share gives earnings - shareholder participates proportionately in the earning of the company. The earning of the company divided by outstanding number of shares gives earning per share or EPS.

These shares are traded in the market. PE ratio is the ratio of market price divided by earning per share. Just as 1000 (in the cow example) was the ratio of trading price of a cow divided with the number of litres of milk it delivered a day, PE ratio is the ratio of market price and the annual earnings of a share. Since both earnings and market price of a share are in rupees we have rupees in both denominator and numerator that get cancelled and we get a number without units; unlike in the cow example where the earning is in litres of milk but price is in rupees and we get the PE ratio of a cow in Rupees per Litre.

What does it denote? It denotes how much the market is valuing the earnings of the company. You will find that FMCG marketing companies (Colgate, HUL) etc have much higher PE ratio than say cement makers like Shree Cement and ACC. Market men rate the earnings of FMCG firms more valuable than those of cement manufacturers. The reason usually is a lot of the earnings of cement makers gets absorbed in investments in new facilities. Unlike the cement makers the FMCG marketing companies need relatively little investment in plant and machinery to grow and thus more of their earnings come to shareholders. Also cement business has markedly more cyclicality than say toothpaste. We will brush our teeth even during recessions but most of construction activities will suffer severe slow down during recessions. Thus cement manufacturers' earnings are less stable than those of toothpaste makers. High PE ratio indicates that teh mix of growth and stability is better for that company.

Thus some sectors have higher PEs and some others have lower PEs. Within a sector low PE for a firm means the market does not think that it would grow quite as well as others. Some people think that low PE denotes under-priced shares and thus a buying opportunity. This is debatable. If it were truly a buying opportunity others would have bought it; if the market is rating its price low it is because its value is indeed low and there is no buying opportunity.

(After all it is difficult to assume that only we have noticed that its PE ratio is low.)

Returns and Log Returns

2011-03-25T01:18:00.000-07:00

Why do we consider log of returns from securities while estimating VaR but percent returns while estimating beta? I received this question in mail a few days back.
First about beta. Many practitioners do use log returns for estimating beta. I believe it causes no harm but gives no advantage either.
In a typical VaR model we assume the returns to be normally distributed. In such cases percent returns are unsuitable. The reason is their asymmetric nature. If the price of an asset goes up from 100 to 200 the return would be 100% but if it were to fall next day from 200 to 100 the return would not be -100% but only -50%. The asymmetric nature of per cent returns can be understood from the fact that negative return at most can be -100% but there is no such theoretical limit on positive returns.
By taking log returns, where return on day t of an asset is ln{P(t)/P(t-1)}, we overcome the problem of asymmetry, as shown in the table:

As the table shows, in case of log returns we get (approximately) the same number, with sign reversed. And of course, log returns allow liberal application of calculus, an absolute necessity as many models assume continuously compounded returns.

More on Merger of Exchanges

2011-03-10T23:45:00.000-08:00

I had posted earlier why I do not favour the ongoing global trend in merger of exchanges. Since then I found an interesting piece in The Economist (10th Feb) that too finds the craze for exchange merger strange. The points made are:
1. “24/7 global pools of liquidity” does not really mean anything but is an excellent chat-up line.
2. Big exchange mergers – NYSE and Euronext, NASDAQ and OMX, two big exchanges of Chicago – have brought only lousy returns for shareholders.
3. Bloomberg’s index of global exchanges remains 42% below its 2007 high (global equities are about a fifth below their peak).
And the article concludes that deal making among exchanges appear to be value destroying.

It considers the recent wave of exchange mergers – LSE and TMX (Toronto), Singapore and Australia - that is being touted as bringing in global behemoths but is sceptical if a global exchange will have any desirable features for participants or for shareholders of those exchanges. It finds the typical arguments that exchanges with different suites of products can cross-sell across an expanded customer base and that sales will be boosted by the prestige a deal creates – attract more listings, at best feeble. It however concedes that joining forces may make it possible to combine the technology and back-office platforms being used by different exchanges, cutting costs.

In a recent post Ajay Shah has examined the cost cutting angle. He argues that an exchange is essentially an IT facility that processes orders to arrive at deals. Since order processing must take place in milliseconds in these days of algorithmic trading and since information can travel only at the speed of light, which is only 300 kms per millisecond two joining exchanges that are thousands of miles apart (LSE and Toronto, Singapore and Sydeney) will surely become inefficient if they share a common back-office platform.

The conclusion: merger of exchanges does not make sense.

Consultants and Inside Information

2011-03-10T02:56:00.000-08:00

Raj Rajratnam, accused of trading in securities on inside information, is going on trial in US. Two Indians are accused of having provided him such information – Anil Kumar and Rajat Gupta, both of McKinsey. Kumar has, reportedly, admitted to having passed on inside information to Rajratnam for money. Gupta has denied having done anything improper. Many believe that Gupat’s name would be cleared after the trial, however, most feel that his reputation has already suffered whatever may be the verdict.

There is an interesting piece in FT. It seems consultancy business to some extent consists of but sharing inside information. Consultants essentially share their knowledge of a client’s competitors when they describe “best practices”.

In 1956 an IBM executive asked Watson, his chief, whether they should share their business information with their consultants, Booz Hamilton. Watson replied that all information must be shared with the consultant, who was just like a doctor. Soon after it the Booz consultant informed Watson that he was joining as president of RCA, then a big competitor of IBM.

Why I do not Favour the Ongoing Global Trend in Merger of Exchanges

2011-02-18T22:28:00.000-08:00

Stock exchanges started as private clubs of brokers and stayed that way till technology made it impossible. Anyone who bought or sold shares before NSE’s appearance will remember how she bought only at the highest and sold only at the lowest reported rates for the day. This by no means was the worst practice, but I will spare you the gory details. Exchanges are no longer “associations”; they are “for profit” limited liability companies. Institutions and individuals with deep pockets are buying chunks of equity in stock exchanges around the world. Exchanges themselves are busy buying other exchanges / selling themselves with an almost unreal gusto. It is speculated that eventually, after this merger phase is over, there will be only a fistful of stock exchanges left across the world; just as we have in credit rating business today. Will that be good for individual savers?

Individuals and institutions are paying huge amounts to buy equities in stock exchanges. Some have plans, no doubt, to sell to the bigger fool; some others, on the other hand, plan to buy and control the exchanges. What could be their incentive? Major sources of a stock exchange’s incomes are transaction charges, subscription fees and listing fees. They also earn substantially from investments. Of these, subscription fee has somewhat less growth potential. Growth in listing fee is directly linked to new firms getting listed, which should be high during bullish phases and low during bearish phases. Transaction charge is more nuanced.

Transaction charge is the charge levied by the exchanges on each trade. Clearly it will grow with the number of transactions. The number of transactions again generally increases in bullish phases. In any phase the number of transaction will depend on the instruments / products available for trading. Globally debt products record higher trading volume than equity products, and derivatives record highest of all. One way of increasing trading volume (and thus transaction charges of an exchange) is introducing new tradable products.

It can be argued that trades will not take place merely because an exchange introduces some new tradable product. I do not think an exchange will introduce a new product on its own or even all of a sudden. Right kind of noises will be made, for a respectable length of time, among the trading fraternity and a consensus view will gradually emerge that the time has now come for this new product which will make the markets more complete, offer such opportunities to the investors in risk management as were not available till now and so on. This will surely dilute the traditional role of stock exchanges, that of differentiating boys from men, whose papers were accepted for listing / trading. I shudder to think what will happen when exchanges, to increase their shareholders’ wealth – no less, chummy with red blooded traders will go all out to introduce new products.

Over last three years the economists are increasingly agreeing to the view that more financial depth and more liquidity of more complex products bring little economic value; they add little to economic growth but enormously to the risk of the financial system. I have argued earlier that CRI’s (credit rating institutions) should be better (read more closely) regulated. In fact I feel CRI’s should not be listed entities. I also feel that stock exchanges should not be allowed to get listed. There should be a cap on the return on equity they can earn (just like power distribution utilities) and am in agreement with quite a few of the recommendations made by Jalan Committee.

The Florist at Bada Hanuman Mandir, Nagpur

2010-12-12T04:29:00.000-08:00

I was at Nagpur for last twelve days, part of a team assigned to an incredibly boring and time consuming job. Finally when it was almost over, a colleague, who has passed out from LIT - a local engineering college - suggested that we should pay homage to Lord Hanuman, the "sankata mochan", (one who delivers you unscathed from crisis situations). Our predicament was such that we readily agreed and drove down to Bada Hanuman Mandir near Lake Phutala. Some of us bought flowers and coconuts from an old lady who had a stall right on the temple entrance. One of us paid her for everyone's purchases and we proceeded one by one to offer our private thanks to the lord on the deliverance that seemed almost certain.

Afterwards we assembled at the stall to collect our shoes and the friend of LIT gave the lady a hundred rupees note. The woman assured him that all purchases were paid for. Our friend insisted that he wanted her to keep it. "Why should I" asked the shopkeeper. "Because I want you to" said our friend. "Why should you want such a stupid thing?" wondered the old one. My friend patiently explained that he had known her since his college days and that her son daily carried my friend's lunch box to his hostel room. The lady was still perplexed; she failed to see any connection between the past and the hundred rupee note and absolutely refused to accept the money, muttering something ominous about my friend’s mental state.

In her situation, perhaps, I would have gleefully accepted the money. I salute her self-respect.

Friday's Spurt in Call Money Rates

2010-10-31T13:54:00.000-07:00

On Friday, the 29th October, the inter-bank call rate touched a two year high indicating acute shortage of liquidity. The reasons ascribed: massive outflows because of Coal India’s IPO, Diwali related cash expenditure by members of public, burgeoning deposits of Central Government’s with RBI.

Public offering of shares by companies normally does not affect liquidity of the system as a whole. Cheques drawn on one bank (the issue subscriber) are credited into another bank account (the escrow accounts maintained with bankers of the issue). Nevertheless large issues can distort the distribution of deposits among the banks: some banks may see their deposits rise substantially at the cost of some other banks’ deposits.

But Coal India’s IPO was not exactly a normal public offering. It was an “offer for sale” by an existing shareholder, namely the President of India. Those getting allotments will not get new shares to be issued by Coal India but the shares being sold by the Government. The sale proceeds of the IPO will not go to Coal India but to the Government. Thus, this public offering will more than merely distort the distribution of deposits among banks; it will also reduce the liquidity by sucking away more than Rs 15000 crores from the banking system.

Normally in an IPO the proceeds are kept in the escrow account till the designated stock exchanges grant permission for trading in the offered securities. I do not know whether this provision also applies in public sector divestment; (SEBI / Government have diluted quite a few other provisions). Since Coal India shares shall start trading, reportedly, on Nov 4, I suppose the permission to trade either already has been granted by the designated stock exchanges on Friday or is likely to be granted on Monday. It implies massive suction of liquidity.

On top of that, there are TB / Dated Securities auctions scheduled for Monday (today). Hardening of short term interest rates is not at all surprising, what is surprising is its sudden rise.

How about impact of Diwali related expenditure? Well, we all withdraw cash for Diwali expenses – substantially more than what we do in regular months. While purchases made through cards will not impact the system’s liquidity, cash purchases will. Cash withdrawn from banks and exchanged for goodies at shops will get back to banks only after a lag and that lag will stretch the banks’ liquidity position.
I am not convinced of the impact of the central government’s balance with RBI. This chart shows the CG’s balances over past few weeks and also over corresponding weeks last year:

Last year, the CG’s balances were well over 30,000 crores from Sep 15 to Oct 09 but call rates were in normal range.

Some other questions:

Did the markets expect it? Yes it surely did. MIBID and MIBOR rates, determined on the basis of a somewhat complex averaging of polled rates every morning surely indicated that those polled expected the rates to be very high. Since the overnight rates on Thursday, the previous day, were nothing extraordinary, the polled rates of MIBID and MIBOR, taken separately, did not indicate anything earthshattering, but taken together they did. MIBID and MIBOR declared on the 29th morning were 7.46 and 7.66 per cent respectively, as against 6.54 and 6.66 per cent previous day. MIBOR was up by one percentage point or by 15% clearly indicating a day of contrained liquidity. What is really remarkable is the spread between MIBID and MIBOR. It was 30 basis points, not seen for almost two years. Mostly this spread has been in the range of 8 to 12 basis points. And whenever it is 30 bps or more call rates have been in double digits!

RBI should have seen it at 9.30 in the morning if not sooner. The CRR adjustment and the second LAF window could have been announced before start of inter-bank trading and avoid the mayhem. (As it turned out the second LAF was hardly utiilised on Friday.)

How long will this phase of constrained liquidity last? Long enough, with IPOs / SEOs of PSU’s lined up to suck out over 20,000 crores and the next date of advance tax payment barely 45 days ahead I think we are moving into a high rate phase that may last this financial year (Mar ‘11).

What is Enterprise Value? And, how should we compute it?

2010-10-29T06:15:00.000-07:00

Should we deduct cash / near cash from sum of equity and debt values to compute enterprise value? Should we add minority interests to market capitalisation and debt value before deducting cash to arrive at enterprise value? These are some of the questions one often comes across.

Enterprise is defined as an organisation created with a commercial intent and thus should be synonymous with a business firm. Enterprise value, then, should be synonymous with firm value. Firm value is the present value of the claims of all stake-holders against a firm. In a simple case where a firm is financed with only debt and equity, firm value is the value of debt and value of equity. If the firm has no non-operating assets, firm value can be valued by estimating free cash flows from operations and discounting those cash flows with an appropriate rate of return (which may vary from year to year) that embodies the returns required by creditors and shareholders of the firm. Since this is essentially the value of operations of the firm, if there are any non-operating assets, the value of such non-operating assets should be added to the value of operations to arrive at the firm value or enterprise value. (Copeland, Koller and Murrin have used enterprise value in this sense in their book)

In short enterprise value is the value shared by all investors, unlike debt value or equity value, which are shared only by creditors or by equity-holders respectively.
However, the term “enterprise value” has acquired a new meaning for last fifteen years or so. It is said to represent the minimum amount one would need for acquiring a firm, free of its debts. In this avatar it is often defined as:

a)Market value of equity + Market value of debt + Minority interests – Cash (Damodaran, see here)
b)Company's market capitalization - cash and cash equivalents + preferred stock + debt + minority interest (JP Morgan’s site, see here).

Investopedia, a popular website too computes EV as market cap plus debt, minority interest and preferred shares, minus total cash and cash equivalents. It explains why cash is deducted: “think of enterprise value as the theoretical takeover price. In the event of a buyout, an acquirer would have to take on the company's debt, but would pocket its cash.” See here. Intestopedia also defines another similar quantity called “Total Enterprise Value” or TEV as Market Capitalization + Interest Bearing Debt + Preferred Stock - Excess Cash. See here. Even this “total enterprise value” does not come close to what I have understood to mean enterprise value, i.e. the value of the firm. I will call the other definition of EV (mentioned above as a or b), as the trade definition. (It does not mean that all academics interpret EV as the firm value; I have seen a few scholarly papers where EV has been taken as what I have dubbed here as the trade definition.)

I am not comfortable with the trade definition, even if it is defined as the minimum amount that would be required to acquire a firm. My discomfort arises from two points:

1. First concerns deduction of cash. The logic given (by Investopedia for instance) that after acquisition this cash will be available for taking out presupposes that the cash is not needed for operations, in other words the operations of the firm do not need any cash and that cash can be removed without affecting the value of operation. This certainly can’t be true for any company. Some companies may have large amounts of “excess cash” i.e. cash balance over and above what is required for operations. This excess cash can of course be taken out without affecting operations but not the entire cash. Most companies may have some other non-operating assets albeit less liquid than cash. For example most (established) companies in India own excess land i.e. land not needed for operations. Will it be correct to deduct the value such excess land may fetch in a distress sale for arriving at EV? After all that value can be taken out from the company without affecting the operations at all once acquisition is complete.

2. The second point relates to adding minority interests. Minority interest is generally understood as a significant but non-controlling ownership in a business entity. Suppose the target firm (whose EV we are interested in) owns 20% of another firm, Z, that too is listed (for our convenience). Trade definition of EV implies that 20% of the market value of Z should get added to the market cap of the target. This pre-supposes that the market ignores the minority interest of the target in Z and thus the market cap of the target does not consider the value of its partial ownership of Z. I think this is being unjust to the market. In a reasonably efficient market minority interest should get reflected in the market capitalization; I however have no hard evidence for it. (I dare say we do not have any evidence to the contrary either i.e. the minority interest is NOT reflected in market capitalisations. There are, however, assertions that minority interests are valued at discount by market.)

In view of the foregoing I am convinced that the practice of adding minority interests and subtracting cash from market cap plus debt value of a firm for arriving at its EV is incorrect. To correctly determine EV one should first value the operations of the firm and then add to the value of operations so arrived all non-operating assets,including excess cash and minority interests.

Stray Thoughts on “WACC: definitions, misconceptions and errors” by Pablo Fernandez

2010-10-26T22:07:00.000-07:00

Re-read today an interesting paper “WACC: definitions, misconceptions and errors” by Pablo Fernandez. The paper can be accessed here .

Fernandez maintains that WACC is an average of two quantities that are very different in nature – one is a cost and the other is a required return, and hence WACC is neither a cost nor a required return. He then lists some errors due to considering WACC as a cost and not remembering the definition of WACC:
1.Use of wrong tax rate (T), which should be the effective tax rate for each year
2.Use of book value weights in computing WACC
3.Use of a capital structure that is neither current nor the forecasted one in computing WACC
4.Enterprise value (E+D) does not satisfy time consistency formulae
5.Considering that WACC/(1-T) is a reasonable return for a company’s stakeholders
6.Use of a wrong formula for debt when the value of debt is not equal to its book value

The errors are serious. He has also given some examples – one from a valuation done by an investment bank which had error #3 and the other from the policy of a country that had embodied error #5. He, however, has not shown how these errors arise from mistaking WACC as a cost.

I cannot help thinking that required return is not all that different from cost. After all what is the interest rate on debt if not the return required by the creditors? If return required by the creditors can be considered cost of debt then why the return required by equity-holders can’t be called cost of equity? I do not find any difference between “required return” and “cost”. The title of the paper, I am forced to conclude, is just a bit of polemic. Rest of the paper has one more valuation example and a concise summary of our knowledge on present value of tax shields.

It is a very informative paper with an unnecessarily misleading title.

Pinpoint Equity

2010-10-17T09:01:00.000-07:00

Came across a query today at Finance 30.com: "I am working on a PFI model and has come across a concept of Pinpoint equity. Can anyone help me understand what is this and how is different from the normal PFI fundings."

This is what I have commented there:
It is not different from equity finance as we understand. The sponsors fund the project company with equity and subordinated debt. They prefer to use sub debt route to the extent possible. There are two reasons:

1. Subordinated debt usually starts earning interest as soon as operations commence. Equity dividends remain far away in future - after debt servicing becomes regular and debt service reserve account is fully funded.
2. Also, equity must, ususally, be brought in at the time of financial close whereas debt is brought in as per a draw down schedule.

Why is their (sponsors') equity contribution called (by some) pinpont equity I do not know. My hunch is given a chance a sponsor would fund the project company with as much of sub debt as it can and only contribute a minimal level of equity - a pinpoint.

I wonder if my hunch is right and shall sure be glad to learn more on it.

Added on 18 Oct: Learnt today, equity investment in a project is called pinpoint equity when it is brought in (entire) before any debt - senior / mezz etc

Introduction to Merger & Acquisition (Illustrated with Piramal - Abbott Deal)

2010-10-15T13:18:00.000-07:00

A few months back I had prepared an introductory note on merger and acquisition, based on the sale of business by Piramal of India to Abbott of USA. I have today uploaded the note on Scribd; can be viewed here.

The notes have also been published at SSRN, available here

Comments welcome

Road and Highway Project Modelling

2010-10-10T02:24:00.000-07:00

Often I receive requests for financial models for road, power plant and other such infrastructure projects. I do not think it is quite right. Now I normally ignore such requests but earlier I used to share my thoughts / feelings about this approach. I stopped sharing my views after an unsavoury experience.

A long time back I had replied to a Palestinian national’s similar request suggesting that he should start building the model and if he stumbled on any specific difficulty I would surely try to help. This suggestion was not taken kindly. He later wrote that there was no shame in seeking help and instead of helping him I had started to “dance”. (I had never come across this use of the verb dance.) I still maintain that when assigned to model a new activity one should not look for a template. A template will not help us grow and only a few people ever dare to think beyond the template.

Some time back I came across this request: “I am looking forward to start building model on road project (BOT-Toll). I don't have an idea where to start. Looking for any guidance.” I liked it because the correspondent was not looking for a model but for guidance. This is what I wrote back. I do not know whether it helped her but it seemed to have bucked her up. I feel with a little bit of modelling skill one can model a Toll Road Project by following the steps outlined here.

First thing is to just start building the model! Understand what is to be modelled and then just do it. It will necessarily be an iterative process. The first version in all likelihood will remain unfinished but it will tell us what questions to ask the concerned people.

For a toll road model, start with construction. Learn how long the construction phase will last - what are the costs involved, what equipment need to be bought for the project and how the management plans to dispose them after the project is over (this will help us decide on depreciation). Understand all the processes in construction and supervision including sub-contracting if any, find the costs of each process and then we can work out a model that will tell us the total cost of construction, and also at what points of time how much funds will be required etc. This will not have any financial costs but will have supervision costs.

Next should be the finance module. Decide on the gearing, get the cost of debt and draw from the construction module details on at what points of time how much funds will need to be brought in. We can then prepare the debt draw down schedule and with the terms and conditions such as commitment fee if any, interest rate, moratorium period and tenure we can prepare loan repayment schedule and estimate annual interest costs for every year till the loan is repaid.

Next will be the operations module. Understand the traffic projections if you have. If you do not have traffic projections you will need to prepare one going as long in future as your concession period at least. You will find plenty of material on preparing traffic projections by looking around, usually the project authority gives traffic projections and the concessionaire needs to validate it. This is extremely important. It can be compared to projecting market size and market share for a manufacturing project. The operations module will of course have all maintenance costs including insurance costs.

Next estimate inflation. This is tricky. I would rely on RBI's announcements and ensure that the model permits the inflation rate to be changed from year to year. I’d use the base rates of toll and the escalation formula given in the contract coupled with traffic projections for projecting cash flows.

Next step is linking all the modules. If time is projected along the rows consistently integrating the modules become very easy else care will need to be taken to ensure that items relating to same period of time fall in same column. In this table we will need to apply the taxes on incomes – both current and deferred. I would then proceed to develop the cash flows statement from this integrated table.

This module will also have the waterfall arrangement for disposal of the net cash flows including the debt service reserve account. This statement will give us all data to compute the project IRR. It may not have ready to use data for computing equity IRR but that can be readily remedied by deducting from net cash flows in each period the cash flows related to debt servicing / repayment. Cash flows statements also give us the end of period cash holding, which is a big help in correctly projecting the future balance sheets. I have noticed that many modellers move from operations statement directly to balance sheets and face a lot of problems in getting the two sides of the balance sheets agree. Some (many) use cash balance or another similar account as a “plug” that is adjusted each year to get a balanced balance sheet. This is indeed a very poor practice. It also disables the model form any scenario testing where the required output is some balance sheet ratio.

It is important that one goes through the Concession Agreement - I know it runs into thousand pages or so but we need not go through the technical details. We must pay attention to risk management and incorporate them in the model, though.

Finally we must ENJOY, there is nothing as exhilarating as the first major model in a new field.

Pakistan's Circular Debt in Power Sector

2010-10-08T09:24:00.000-07:00

Some time back I had a lively exchange with HMA, a Pakistani national on Pakistan’s “Circular Debt” problem. Of late I find a few people from Pakistan chance upon this blog everyday looking for circular debt. I feel the discussion with HMA may be of help. I am reproducing the discussion with minor changes:

HMA: What would be the best strategy to overcome this long lasting problem? Suppose you were the Finance Minister of Pakistan and facing the same position and you have to clear the entire circular debt that is hovering around PKR 91 billion. (Previously it was claimed that circular debt problem will be eased through reducing the subsidiaries and increasing the tariff. Subsidies were reduced and tariff was increased but after a few months the same situation aros again and now we are paying higher tariff and ultimately the inflation is also hurting us.

SS: I am not fully acquainted with all the facts. As I have said earlier, to me it does not appear to be a case of circular debt as the circle is incomplete. My knowledge is based on published reports available to me in Mumbai.

Suppose, the circle is indeed complete. In that case, handing over cheque to one party and expecting to receive it back from the last one down the chain would be naive. A more robust mechanism would be to nominate a commercial bank as a central agency (temporary - for this purpose only) that would examine all the links and after assuring itself of the circularity, will open special current accounts for this purpose only.

All claims and liabilities forming part of the circular debt problems will get assigned / novated in favour of the central agency, who will credit / debit the special accounts accordingly. If the claims are properly identified (that they do indeed constitute a circular debt problem) all special accounts will have zero balance at the end of the day, with all claims satisfied.

You might have noticed, what I am suggesting is a typical stock exchange clearing house arrangement. These clearing houses consider the claims and liabilities of scores of brokers and after cancelling out circular claims determine net liabilities / claims of each broker.

It appears simplistic, it indeed is; but if we have real circular debt issue, I think it should be an effective and inexpensive way of tackling it.

I, however, feel we do not have a circular debt problem. The problem seems to be the one we have in India - power theft, inefficient (corrupt) distribution companies and non-payment of power charges by many of those consumers who are not stealing. Tackling it, like in India, will require political grit.

HMA: The government had earlier formed a company named Power Holding Company Limited for this purpose and all liabilities were transferred to this company and as suggest by you but it failed to resolve the problem. In my opinion it is because mere book entries for clearing debt is not a practical solution and companies needed cash for Working Capital requirements. However, they are coming up with a new strategy this time and according to different media reports, the govt. has decided to take the following steps:

- Outstanding electricity bills of all the govt. institutions, including provincial governments would be adjusted from their accounts. Ministry of Finance would adjust all outstanding amounts of govt. institutions from their respective accounts.
- Federal govt. will inject PKR40bn into the power generation system through budgetary resources which will trickle down from IPPs to the E&P companies.
- Additional Term Finance Certificates (TFC) worth PKR12bn in addition to two previous TFCs worth PKR85bn each, will be issued by the govt. to aid the cash starved energy sector.

Your views will be highly appreciated

SS: I tabulated the details of the receivables / payable positions of each agency (sheet 1 of the attachment) and derived the net position. In a true circular debt situation each agency would have net position of zero.
The net position emerges as under:

Three agencies, namely KESC, some private sector firms and governments (Fed +State) account for 90% of the dues of those whose net position is positive i.e. consists of receivables. I understand KESC is in power generation, transmission and distribution. It would be interesting to have a look at its accounts. If they give activity-wise break up, I am sure we will find cash is being lost in distribution - theft and non-recovery of dues.

The remedies are
a) Toning up of KESC;
b) Strict recovery from private sector; and
c) Issue of bonds or increased taxes (by government0.

The first would be resisted by KESC unions, the second by those powerful private sector groups that manage to run such huge bills and do not care to pay and the third by everyone!

As I suspected earlier it will call for a lot of political determination to tackle it. Government dues can be of two types: a) Local governments (districts and municipalities - perhaps railway?) are not paying in full; b) Government is unable to pay the subsidy amounts for concessional / free power supplied to special class of consumers. These can only be remedied by suitable policy amendments. But here again what is actually required is political grit.

HMA: It is clear that the main culprit is KESC (the power distribution company and the problem started when GOP sold it to a Private group - Abraaj group which now holds around 74% stake in it. Back in 2008, they took over the company and started running it solely for generating immediate cash flows. They stopped payment to oil companies for their furnace oil consumption and the load shedding got worst in the city of Karachi, which is the financial hub of Pakistan. Then came politics and everybody shouted WAPDA to allocate further units to KESC to meet the demand of the citizen of Karachi. By the time KESC started recovery process and provincial govt and other govt related offices came under fire and the receivable of these companies are linked to the payment to oil companies and WAPDA (for extra unit purchase from them to meet the demand).

The situation get worse day by day and WAPDA stopped payment to IPPs and IPPs stopped payment to oil companies and they stopped payment to Refineries and Refineries stopped payment to Oil and gas companies.

As on Dec 31, 2009 (from July - Dec ) KESC own unit generation is 4,076 where as Unit purchased 3,970 (millions Kwh) and total units avaiable for distribution is 7,740 where as unit billed (5,055) so the transmiision and distribution losses are 34.68%.

It seems your first suggestion is working like this and I hope this will improve in coming months after some major development has been taken by Abraaaj group.

Your second suggestion will get implemented like this: Outstanding electricity bills of all the govt. institutions, including provincial governments would be adjusted from their accounts. Ministry of Finance would adjust all outstanding amounts of govt. institutions from their respective accounts.

And your last suggestion will get implemented in this manner: Additional Term Finance Certificates (TFC) worth PKR12bn in addition to two previous TFCs worth PKR85bn each, will be issued by the govt. to aid the cash starved energy sector.

But this all need an efficient way to implement and linked with the seriousness of the Government which is till now missing in this case.

Covered Calls

2010-09-26T10:24:00.000-07:00

Two days back my former physics lecturer mailed me; inviting my attention to a recent write up on “covered calls” in the ET he wanted me to blog about it; here I am. First some lines about basics of options.

Options are quite common in real estate. Buyer of a property usually deposits a small amount with the seller and gets the right to purchase the property till a certain time at a certain price. If he does not buy within the period he forfeits the deposit. The deposit was thus the price of the option he had obtained to buy within a certain period at an agreed upon price. The vendor of the property had sold the option and received the deposit as the price of the option he had sold.

Financial markets have two types of options – call and put. Call options give the option-buyer a right to buy an asset within (or on maturity of) a certain period and put options gives the option-buyer a right to sell an asset within (or on maturity of) a certain period. The asset on which options are traded is called “the underlying”. Again put or call can be of two types. In European options the option holder can exercise his right only on maturity of the period and in American options the right can be exercised anytime during the period. In India options with indexes as underlyings are generally European options and those with individual shares as underlyings are American options.

The seller of a call option carries the obligation to sell the underlying at an agreed price and for carrying this obligation the option seller receives a premium up front, just as the property vendor does in the real estate example. Suppose, I sold call option on SBI, expiring October 2010, agreeing to sell SBI share at 3300 (called strike price) on 23 Sep 2010 for Rs 68. I would receive on Sep 23 the option price or the option premium Rs 68 and in return I would be obliged to sell SBI share to the call holder at 3300 on any day (this being an American option) till October end. On Sep 23 SBI closed at Rs 3139.45. The call option I sold was out of money i.e. option holder would have lost money if she had tried to exercise her option immediately – I would have merely bought SBI from market at 3139.45 and sold her at 3300, the strike price. But I am exposed, as I am obliged to sell SBI at ₹3300 till October end and if SBI prices go beyond 3368 I will lose on the deal. (If SBI remains below 3368, at say 3350, I will buy from market at 3350 and deliver at 3300 losing 50 but gaining over all 18 from the deal as I have already received 68 as option premium.) My exposure is described as “naked”. If, however, I buy SBI today at 3139 I get covered. If SBI goes beyond 3300 and I am to make delivery to the call holder I do it with the SBI shares bought at 3139.45 and in the process I also make ₹169 (the difference between 3300 and my purchase price 3139)!

Is it a win-win deal for me? It cannot be because there are so many people out there, much smarter than I am, they would all have been selling Oct calls on SBI and buying the shares in the spot market, in the process decreasing the premium on call and increasing the spot price of SBI.

A covered call just covers my exposure to call I have sold. It protects me against any rise in SBI price. It does not protect me against any fall in SBI price. My profit (on X Axis) from selling covered call (i.e. selling call and covering myself by buying an SBI share) and from only buying a share in SBI, at different prices of SBI (on Y axis), can be seen in this graph:

It is seen that a covered call (CC) seller has limited upside (will gain only 228.55 irrespective of the price of SBI) and has slightly smaller loss if SBI falls because of the call premium received upfront. It makes a covered call like a put option. In put option the option seller is obliged to buy from the put holder the underlying at a predetermined price. Suppose I sell a put on SBI at 3300 maturing at Oct end and I receive X amount as option price today. If SBI goes beyond 3300 on Oct end the put holder will not require me to buy SBI at 3300 from me. But if SBI falls he surely would ask me to buy at 3300. Thus my gain on SBI’s rise is limited to X but my downside is not definite, it will be 3300 - Spot price on Oct End + X. It can be seen that a covered call is but a put.

Is it a very good investment strategy? In fact I read somewhere, attributed to Chicago Board, that a covered call is even better than just investing in stock. I do not know but I firmly believe that unless you are in option trading for living you can rarely do a more stupid thing than selling a put. In fact other than option dealers, no one should ever sell options call or put; naked or covered; European or American. Buying options is different, buyer knows the maximum loss he can incur (the option price paid up front), but selling is not for boys.

The ET article was not describing boys either. It talked about HNIs. All I can say is some smart broker / bank has persuaded some rich guys to take risks without adequate upside potential (remember gain limited to 228.55 in the example irrespective of rise in SBI). Those who believe that the rich and mighty know what they do just have to go back three years to see how the smart corporates of India burnt their fingers by holding red hot forex derivatives that the bankers had assured were the coolest things going around.

FSA Fines Investment Advisors

2010-09-22T06:55:00.000-07:00

Financial Services Authority of the UK fined one firm and two individuals, one of them was partner of the firm and the other was from another firm, for giving unsuitable investment advices to their clients. The law firm Thorntons Law LLP that also provides investment advice to its customers under the brand name Thornton Investment Services has been fined GBP 35,000 and Michale Roydon, one of its partners has been fined GBP 10,500. Read the FSA press release here .

I wish they would take up the rating agencies too but I am sure the rating agencies will threaten to downgrade UK’s rating in retaliation.

Impact of debt financing on Project IRR and equity IRR

2010-09-21T08:27:00.000-07:00

Since IRR is estimated with initial outflow (assume that it takes place only in one period, the first) and free cash flows of all subsequent periods; and as free cash flow to the firm (FCFF) is not affected with interest payment and debt repayment outflows, it is generally believed that inclusion of debt will have no impact on a project’s IRR. But in practice inclusion of debt generally changes the project IRR, albeit marginally. This is because inclusion of debt increases the cost of the project because it adds interest during construction, which is not there in case of all equity funding of a project; thus initial outflows goes up. Interest during construction is generally capitalised and capitalisation of interest also implies higher book value of assets and thus higher depreciation and consequently lower taxation. In jurisdictions with very low or no taxes (I am told that in some Middle Eastern economies only tax, Zakat, is 2.5%) the tax benefit will be minimal and replacement of equity by debt will decrease the project IRR; impact of higher initial outflow on account of interest during construction will outweigh the impact of lower tax outgoes in subsequent years.

An interesting case came my way yesterday. I received a project model with a request to check it for errors. The modeller had submitted it to his superiors who played around with debt equity ratio and found that at 100% equity funding the equity IRR (and project IRR, naturally) was coming to 11.41% and with 50% debt funding the equity IRR improved only to 11.93%. Convinced that the model was faulty they returned it to the modeller and asked him to rectify it promptly or …

My other preoccupations did not allow me to go through the model meticulously. It had some obvious minor errors which were promptly rectified but those errors had no impact on IRR computation. I tried to explain the modeller why inclusion of debt improves equity IRR and the factors this improvement would depend on. This is what I mailed to him, (more or less):

1. Debt financing improves equity IRR because debt servicing (payment of interest) is tax deductible. Inclusion of debt increases equity IRR because of tax shield which is interest payment * Tax rate. In the instant model tax (Zakat) is only 2.5%. Thus tax saving dues to debt is rather insignificant.

2. Inclusion of debt also improves equity IRR as cost of debt is generally substantially less than project IRR. It implies that cashflows available to equity, though lower compared to what would be available in all equity funded case, does not come down proportionately i.e. initial equity investment falls more than the subsequent cash flows to equity. In the instant model, this too was not very effective as the project IRR (on all equity basis) was only 11.41% and cost of debt (pre tax) was 6%. Had the project been more attractive (IRR in the range of 20%) replacing 50% of equity with debt (IRR 6%) would have increased equity IRR considerably.

Beta of Debt

2010-09-06T06:41:00.000-07:00

This question was posted by a finance student sometime back on Finance 30.com: “In most cases I have found that the assumed beta of debt is zero, as its value is negligible.... but i don't know how this conclusion is arrived at without calculating it. Can anyone help me in calculating the beta of debt?”

Since a number of students face this issue every year I am reproducing what I had replied there:

Beta of an asset is the ratio of covariance of the asset’s return with the market return and the variance of market return. If debts of the firm are traded, beta of debt can be estimated by regression just as we do for estimating beta of equity.

Analytically it can be seen that beta of good firms' debts should be quite low - not zero really. Let us consider a plain vanilla debt that offers to pay a fixed coupon at quarterly intervals and repay the principal at the end of its tenure. Total return from holding this debt over a period will be: (value of debt at the end of the period - Value of debt at the beginning of the period + Interest received during the period)/ (Value of debt at the beginning of the period).

If the issuer firm is good, i.e. the debt is highly rated, it is expected that it would not default on coupon payment. If the issuer's credit quality does not change the discount rate for valuing the debt will depend on the interest rate structure on the date of valuation. If the interest rates go up, the value of the debt would come down, and if other things do not change (ceteris paribus) the market should also come down as interest rates go up. Thus there should be a positive relationship between return from market and return from holding debts, and beta of debt, of even good companies, cannot really be zero.

Having said that, the fact remains that equities fluctuate a lot more in value than debts. This makes the positive correlation between equity markets (and we generally take a well diversified equity index as proxy for CAPM market) and debts rather weak. If teh covariance of teh debt's return with market return is low, debt beta will be close to zero as the variance of market return is (visibly) high.

Betas of debts can also be estimated direct from the CAPM equation without going through regression. CAPM equation is:
Expected return from an asset = Rf +Beta of asset*Rp;
where Rf and Rp are risk free rate and risk premium respectively. The equation can be rearranged to give:
Beta of an asset = (Expected return from the asset - Rf)/Rp
Since expected returns from debt instruments (Rd)are known (unlike those from equities) beta of debt will be given by (Rd-Rf)/Rp. If the debt rating is very high (read AAA) Rd-Rf will be very small and (Rd-Rf)/Rp can be approximated to zero. It will never really be zero as in that case Rd must equal Rf, which is never the case.

The nature of expected return from debt is one sided - if the company goes bust the creditors lose but in boom they do not get more than what is contracted. This limited upside potential means the return form debt would not be high when the market gives high return. This makes use of CAPM equation / beta for determining return on debts unjustified.

Beta changes over time, and how!

2010-09-06T05:22:00.000-07:00

Beta is a popular and an useful metric. It is a neat number and in one number we get an idea of the risk the asset is going to add to a well diversified portfolio. The important points to consider while estimating beta of any assets are:
1. An appropriate proxy for the “market”
2. Return periodicity i.e. do we consider daily return or weekly return or monthly return etc
3. Period of analysis – over one month, one year, five years ...
I go for weekly returns over a two year period. This gives a data set of about 100 which is just about large enough for a meaningful statistical analysis. Daily returns, I feel, have far too much of “noise”. I also feel that in these fast changing times analysis over a long period (say five years or longer) is not very appropriate as the risk profile of the firm may suffered significant changes over the period. Many others feel that one should take monthly returns over a five year period for a better estimation. I will say it is good that we have diversity.

If risk profile of a firm changes over time, so should its beta. I have estimated rolling betas for a few firms in the past and sure enough betas have varied over time but not really in any dramatic fashion. A typical example is SAIL’s beta over last seven years:

This is the latest such estimation I have done. SAIL (Steel Authority of India Ltd) is a public sector steel manufacturer. Betas are calculated with NSE index NIFTY as the proxy for market, weekly return periodicity and over two years.

SAIL’s beta has risen significantly from 0.9 to 1.4+. But compare with a case reported by Bill Rempel here :
Simon Property Group (SPG), a REIT, currently shows a Beta of 1.75 to the S&P 500 (it is a member). In a spreadsheet I confirmed that the slope of the regression line for the prior 36 monthly adjusted (incl. dividends) price changes of SPG to SPY (an ETF tracking the S&P 500 index) is 1.75, so that is the basis (monthly changes over the last 36 months) of their Beta calculation.

I then calculated a rolling 36-month Beta for SPG from Jan 1997 (3 years after it started trading) through today. 45 of the 164 observations showed negative Beta. Most of these instances occurred during the Tech Bubble (in which REITs didn't participate) and during the following Bear Market (in which REITs didn't participate due to the then-starting Real Estate Bubble).

It was not until Oct 2005 that SPG's Beta exceeded 0.50. An analyst naive enough to see Beta as a property of stocks, or foolish enough to think that correlations hold relatively constant, would have been damaged if they chose to try and make money on those assumptions with SPG, as by Apr 2006 SPG's Beta to the S&P 500 began to consistently exceed 1.00, generally exceeding 1.40 since that time.

Bill concludes: Be reminded that "beta" is merely an observation about a stock's return over some sample period, and not a permanent, immutable characteristic of that stock. "Betas" change over time, sometimes very rapidly, over time.

Micro-finance and High Interest Rates

2010-09-04T06:27:00.000-07:00

The proximate cause of this post is an editorial in Phalanx . I am thankful to Ratnakar for posting the link on FB.

Some time back I had linked Ketaki Gokhale’s article in WSJ. Gokhale points out how moneylenders are being repaid in Mehboobnagar of AP that has highest concentration of micro-lenders and moneylenders, by availing loans from micro-lenders. Gokhale also lamented an unrealistic repayment term of micro-lenders that demand the repayment to start from next week itself. As Gokhale rightly says most ventures do not generate instant profits. In such cases the micro-borrowers borrow from moneylenders to keep their track records clean at micro-lenders.

An old friend of mine, credit officer at a bank in Gulf, after reading Gokhale's piece sent me a longish mail describing how the regular laws of modern finance break down at the micro-finance level just as the laws of Newtonian physics break down at the atomic and sub atomic levels. (Emphasis mine but the sentence is almost a direct quote.) This is the first example he has given:

An urban vegetable seller (Cash to Cash cycle of 1 day)

Looking at it one might agree that MFI’s are quite justified in charging high rate of interest. After all they charge merely 3600% from a business that earns 21600%. What my friend has missed is the cost of manpower in a manpower intensive business. Minimum wages is about 150. If a person is undertaking a business in which one his fully engaged (and perhaps that person’s spouse is part time engaged) it is reasonable to suppose that the labour cost of the business would be about 250 per day. If we load that then we find that the business’s return before interest, depreciation and taxes is 3600% and it needs to pay interest @3600%, as illustrated in the table.

Let us take an example from organized sector. Infosys is in a manpower intensive business.

Infosys Technologies (FY 2010, calculated / extracted from Annual Report for 2009-2010)

Do we say that Infosys enjoys 85% profitability? We do not say so because we feel that employees of Infosys, good folks like us, are entitled to their rightful wages and those wages must be deducted from the business’s earnings to work out the business profit. The vegetable seller, poor sod, should get what is left after selling. He is not incorporated; his business cannot have an existence other than his own and why complicate the matters with things like wages for these guys?

Worst of all, if someone talks about the wages of such people being other than their business incomes, surely, that guy must be a wretched communist and has no business talking about micro finance the newest capitalist toy.