Infosys Technologies INFO IN
January 01, 2005 - 11:34am EST by
2005 2006
Price: 2,089.00 EPS
Shares Out. (in M): 0 P/E
Market Cap (in $M): 12,700 P/FCF
Net Debt (in $M): 0 EBIT 0 0

Sign up for free guest access to view investment idea with a 45 days delay.


My recommendation is setting up the Infosys Technologies share class arbitrage.

The $12.7bn capitalization for Infosys I have listed in the idea specifications is its capitalization in the Indian market. The capitalization in the US market as is $18.7bn.

The reason for this discrepancy is that the ADR is trading at just under a 45% premium to the underlying ordinary share equivalent. Naturally, one can freely convert ADR’s to ordinary shares (nobody would) but not the other way around.

The ordinary closed in India under ticker INFO IN at an Indian Rupee price of 2089 INR (INR being Indian Rupees). At a rupee dollar exchange rate of .0229 dollars to the rupee and 1 ADR/ordinary share, that equates to $47.83. INFY, the US ADR closed at $69.31.

My recommendation is to set up the share class trade, buying the Indian ordinary and shorting the ADR. As for the ratio, I will address that in further detail below. For the moment, I suggest trading it at a range between roughly equal dollar amount on the long and short to 5-10% net long market value.

Infosys is not the only Indian company where there is a big differential, and others are interesting as well. There is Wipro, Satyam, HDFC Bank, State Bank of India, ICICI Bank, plus other existing and future ones. As a general principle, over time, as the ADR’s and ordinaries are identical economic instruments, I believe they will inevitably gravitate to parity. This year alone, the INFY premium has ranged from 34% to 68%, and the range has been wider than that over time.

But I feel that the timing is right, and that INFY is the best name in which to invest.

Such market inefficiencies can exist for a long time and trade anywhere until their likely resting point of parity is reached. In addition to the lack of convertibility as listed above, there are a number of reasons for the difference in prices. 1) There is a limited supply of the ADRs, currently only $1bn worth of INFY at US dollar prices. This is a relatively small float in relation to the size of Infosys ($18bn) and the amount of capital available to invest in the company. For many investors, the US ADR is the only way to gain access to the company so the ADR has a scarcity value. Until recently, there were also foreign ownership ceilings in India for the company. Additionally, it is difficult, to own the local shares. Consequently, the ADRs and ordinaries do functionally trade to a certain extent as somewhat unrelated securities with different potential investors. A Goldman report earlier this year actually suggested that the ADRs should trade at a premium to the underlying ordinary shares as the cost of capital is lower in the US.

As for buying the ordinary share, in all Indian companies, to buy them directly, one needs to register as an FII, Foreign Institutional Investor. It is not easy to do, and it seems impossible to do if you are a Hedge Fund…. for now. Suffice to say that SEBI, the Securities and Exchange Board of India (their SEC) wrestles with the role of hedge funds in the market. Without going into too much detail, the issue has to do with the concern that Non-Resident Indians (NRI’s) would use their own created hedge funds which maintain their anonymity to funnel money back into the market, potentially in a manipulative fashion. So for all practical purposes, for the moment, the only way to buy the ordinary share is on swap (P-Notes as they are called in India, where P stands for Participating). This further weeds out potential buyers of the ordinary shares as not everybody can buy on swap. Finally, earlier this year, SEBI essentially ruled that in order to buy p-Notes, the fund, its GP, or Advisor had to be registered somewhere, even if only in its home jurisdiction, further weeding out potential buyers for ordinary shares. In short, there is a form of administrative geopolitical risk to owning the local shares as SEBI could always change the nature of the playing field.

For those of you who wish to go into greater detail or track over time, the following is SEBI’s web site as well as a key citation (which often can not be clinked on to access).

If the latter does not work, you can to the main site, go to legal framework, then regulations, scroll back to January and February and look at the FII information.

The following publications track the Indian news flow relatively well.

The above factors seem to explain why the local and ADR shares have and continue to trade at such different values. But the situation is changing both on the short and long side of the trade.

First, on the short side, Indian companies are planning more ADR issuances. INFY is one of those companies that have announced one (more on this below). The companies are beginning to want to take advantage of the valuation discrepancies. Additionally, given the mechanics of the offerings, the offerings permit insiders to sell stock at a premium price. As managements of the companies are becoming more financially sophisticated, we believe that this trend will continue. Having a greater supply of ADR’s will tend to reduce the scarcity of the ADRs by increasing supply, which should help to reduce the premium. It also reduces the risk of a short squeeze and the premiums blowing out to further extreme levels (as was seen in the past).

Additionally, as these companies have greatly increased their market caps over the past two years, the risk of huge moves to the upside on the short side is reduced. It is harder for an $18bn company with real earnings to move to an extreme valuation as opposed to an earlier stage and smaller capitalization company.

On the long side of the trade for the local shares, it has been (although not in a straight line) and will likely become easier over time for foreigners to buy the local shares. Foreign ownership limits have been and continue to be raised. The government is working on starting to make it easier to get registered as an FII and has stated informally that over time that it expects to allow hedge funds to register. Big picture, over time, as in many other developing markets, financial markets will increasingly open. As does any developing and rapidly growing country, the need to drive the cost of capital as low as possible seems likely to inevitably lead to a more open market-- although this is clearly one of the risks on the long side of the trade and potentially the spread trade as well.

While it is unlikely that the shares will be fully fungible for quite some time, I believe that it will become increasingly easy to buy the local shares, and that as this happens, demand for the local shares will increase relative to the ADRs. I will point out that as an offset, while the long-term capital gains in India is 0, there is a 10% short-term capital gains tax. While one gets a credit on this, it is a disincentive to investing directly in India. There is no tax on the swaps as the counter-parties that own the securities are domiciled in Mauritius, which has a tax treaty with India.

Additionally, if one steps back and considers the situation from the vantage points of the company managements, particularly for information technology companies that have substantial customer bases, this kind of market price discrepancy is inefficient from a capital market point of view and does not necessarily project the kind of image that these companies want.

Turning to Infosys, this company has a blue-chip, world-wide customer base including very substantial business with US Fortune 500 corporations. It wants to raise capital as efficiently as possible and be perceived as an equal to its customers. The share structure is not consistent with this and should over time lead management to push for convergence. Reflecting this, as well as its apparent desire to join the NASDAQ 100, INFY is about to issue an extra $1bn of ADR’s doubling the float. Unless the ADR continues its substantial appreciation, it will likely have to issue another $1bn over the next 18 months to 2 years to get into the index.

Of all the companies, Infosys has been and seems to continue to be the company that is most aggressive in managing its capital structure. It is doing an issuance now, and it also did one in July of 2003. No other company has sold this much ADR value. Consequently, this company also has the biggest ADR float, and it also trades the highest volume. Combined with another potential issuance out a year to 18 months, this seems to be the safest ADR from the vantage point of not getting squeezed on the short side, and it is also the easiest to trade.

Neither Infosys’ 2003 offering nor this offering will increase the number of shares outstanding. What did and will actually happen is that ordinary holders in effect tender their ordinary shares into the ADR offering in exchange for ADRs and then sell the as ADRs through the offering. The end result will be more ADR and fewer ordinary shares outstanding, a good result if one is long ordinary shares and short ADRs. (Given that Infosys’ management is among the more financially sophisticated, and these offerings provide an opportunity to sell ordinary shares at a premium, management is likely motivated to continue these kinds of transactions.) Consequently, this trade could be timely since, if one owns the local shares before the Q1 offering, one can tender these shares into the offering. Hopefully, this result is not fully discounted by the market.

The mechanics of the offering are unclear and may not yield the same pro-rate for all investors. 6% of the local shares will be exchanged. In 2003, your pro-rate was determined by the total amount of your holdings, not what you tendered. For those owning shares on swap, it is the total holdings of the counter-party that count (assuming it is done the same way again this time). As such, it is not clear what one’s pro-rate will be, and it could be less than 6%, an atypical outcome.

Of course, some shares will certainly be sold. It is for this reason that I recommend setting up the trade slightly net long capital. Share class trades are generally set up somewhere between one share short for each share long (a net capital short position) to an equal dollar weighted, the latter being the pure way to invest in the percentage change of premium thesis. Given my view that Infosys has excellent long-term prospects, were there not a tender upcoming and then certainly post tender, I would suggest setting this trade up on an equal dollar basis. Historically, these trades tend to make the most money in weak tapes, and this has certainly been the case for the Indian ADR/ordinary trades. Stated differently, the ADR and local tend to move in the same direction even in the short term, but given the smaller float, the ADR tends to have bigger percentage moves. As the spread is volatile, I tend to like to trade the volatility for some portion of my position while not disturbing the core investment.

Infosys provides a full range of IT consulting and software services; essentially, it is one of the major outsourcers. INFO IN is trading at 21x projected 3/06 earnings and the INFY is trading at a nearly 32 PE. The company is growing rapidly and is substantially cash-flow positive. The one danger in setting the trade up and not being net short is the possibility that the stocks go down substantially, in which case no money will be made. The company is extremely well run and in a great space. As such, I don’t think there is a major risk of a large decline. On the other hand, I am cautious about being too much net long here as labor market competition in India is putting extraordinary pressure on wages, and the rise of the Indian Rupee against the dollar is also a headwind.

To answer an obvious question, Indian companies that buy back stock cannot reissue them for at least 12 months, thus precluding the company selling ADRs and buying back ordinary shares. (I would expect that over time, the companies will give more thought to this Western type of financial engineering, but based on conversations, it is not something to expect in the near-term.)

To summarize, here is an opportunity to buy and sell the same economic instrument at a price difference of about 45%. There is no apparent “economic risk” to the trade (although never say never). The risks are market, timing, legal, and technical. With patience, perhaps a lot, a substantially favorable outcome seems highly likely.


upcomming ADR offering, expected convergence to parity over time
    show   sort by    
      Back to top