|
|
|
|
IT
Firms and Financial Markets: A Changed Relationship |
|
Jul
14th 2008, C.P. Chandrasekhar and Jayati Ghosh |
|
India's
stock market has lost its lustre for those counting
their wealth in terms of the value of the paper assets
they hold. Between the end of October 2007 and early
June this year, the Sensex has fallen from the 20000
level to around 15500, or by close to 23 per cent. That
decline implies a substantial loss of paper wealth that
hurts most those who bought into the market at its peak.
But not all financial investors need be losers of this
magnitude. What is noteworthy is that if we look at
the Bombay Stock Exchange’s Information Technology Index,
the decline in its value between October 29, 2007, when
it recorded its previous high of 4712.5, and 3 June
2008, when it touched 4542.8, was less than 4 per cent.
This difference is remarkable indeed, but is partly
a result of the fact that the IT sector has only marginally
reflected the unprecedented boom that was witnessed
in India’s stock markets since 2004. As Chart 1 makes
amply clear, when the Sensex was rising rapidly and
soaring above the levels it had reached before 2004,
the IT index recorded much lower rates of increase and
stood way below the peak level of 8613.5 it touched
on 21 February 2000. The highest level it touched in
the period since 2004 was 5611.3 on 19 February 2007.
A consequence of this has been the widening distance
between the graphs reflecting movements in the Sensex
and the Information Technology Index.
Thus, the experience during 2004 to 2008 was the opposite
of that witnessed during the stock boom at the turn
of the millennium, when the stocks of IT companies induced
a degree of buoyancy into the market, with the rise
in the Information Technology Index proving to be much
higher than that of the Sensex. During those years,
IT companies came to dominate the market. This comes
through from the following quote from a report in this
paper in its issue dated 2 January, 2000:
Chart
1 >>
"Though corporate India has undergone many
changes over the last 10 years of reforms, the year
1999 would undoubtedly be one of the most decisive periods
in terms of accretion to shareholder wealth. Over the
last one year, the shareholder wealth (represented by
the market capitalisation of the BL 250 Composite Index)
has increased by 90 per cent from Rs. 3,42,553 crores
to Rs. 6,51,471 crores. Though the top 25 companies
still account for around 64 per cent of the total market
cap, the composition has undergone significant changes.
There were just four information technology companies
among the top 25 in the beginning of 1999 with a total
market cap of Rs. 24,563 crores. Whereas at the end
of the year the number had increased to seven with a
total market cap of Rs. 1,65,747 crores. While the four
companies previously accounted for just 7 per cent of
the total market cap, the 7 companies now account for
25 per cent. The information technology companies have
pushed their way ahead of companies belonging to FMCG,
pharma, commodities and economically sensitive sectors.
Over the last one year, the weightage of major information
technology stocks (in terms of market cap) have gone
up quite significantly. While that of Wipro has gone
up from 2.48 per cent to 8.87 per cent, the weightage
of Infosys Technologies has jumped from just 1.39 per
cent to 7.34 per cent."
One mistaken conclusion that was derived from this trend
was that the rapid growth and the immense promise of
the IT sector had made it one where share values have
come to reflect the true worth of companies. This in
turn was seen as paving the way for the "ownership
economy" where what an individual owned rather
than what she earned was the right indicator of success.
By way of evidence, reference was repeatedly made to
the early adoption by the information technology industry,
led by Infosys, of employee stock options (ESOPs). One
journalist writing on Infosys at the end of 1999 (India
Today, 8 November, 1999) waxed eloquent thus: "Of
the 4,782 employees, lovingly addressed as "Infoscions",
1,667 now hold ESOPs. Of them, 1,376 have stock valued
at over Rs 10 lakh each. Among them, the total number
of the jeans-clad "crorepatis" is a staggering
412. And 97 of them have left the Rs. 1 crore milestone
far behind, having become dollar millionaires (Rs 4.3
crore-plus) by age 30-35. No other company in India,
not even Infosys' software rivals Wipro Ltd and Satyam
Computers, has shared its wealth with such a large number
of employees."
There were a number of features of the IT industry's
romance with the stockmarket that was missed in this
type of analysis. The first was that stock prices were
at levels not warranted by fundamentals. As Chart 2
shows, at the peak of the boom price earnings ratios
of the most successful IT companies were clearly at
levels that indicated that these stocks were overpriced,
possibly because they had been chosen by speculators
who were in essence manipulating markets. Not surprisingly
that boom did not last. What is more since then price
earnings ratios have been, even when high, within ranges
that do not reflect "irrational exuberance".
Chart
2 >>
A second feature of the IT industry’s stock market performance
was that a few companies accounted for much of its dominance
over the market. Thus, as noted earlier, a few companies
accounted for a large share of aggregate market cap,
with three or four companies hogging the stock market
show within the industry. Consider the much-publicised
Wipro story. On the 3rd of January 2000, Wipro's share
price ruled at Rs. 2,809. In a bull run that began around
the middle of the month, the share price climbed almost
continuously to touch Rs. 8929 by 18th February. In
a world where stock values are increasingly used to
value individual wealth, this close to 220 per cent
increase in the course of a month had placed Premji,
who owned around 75 per cent of Wipro stock, among the
world's richest people.
There were other such stories, even if they were less
dramatic than this example of a move from a position
of relatively puny wealth by international standards
into the ranks of the world's wealthiest. This easy
movement to the apex of the wealth pyramid added one
more cause for celebrating India’s IT success. What
was disconcerting was that market capitalisation, or
the value of a company computed on the basis of the
price at which individual shares of the company trade
in the market, was increasingly replacing real asset
values as the principal indicator of company size and
worth. The top 50 or 500 are now determined by many
analysts not on the basis of asset value but market
valuation. This is part of a larger disease which assesses
the size (and ostensibly, therefore, maturity) of India's
stock market based on total market capitalisation.
The consequence was a picture of dramatic change. In
1990, before the reforms began, total market capitalisation
in the Bombay Stock Exchange (BSE) was less than Rs.
100,000 crore - a level that the market capitalisation
of Wipro alone crossed in early 2000. Aggregate market
capitalisation in the BSE reflected an average increase
of 100 per cent a year during the 1990s. This was not
only taken as suggesting a dramatic rise to maturity
of the stock market, but as reflecting economic buoyancy,
even though it conveyed a completely different picture
than that provided by GDP growth, which averaged around
6 per cent compound a year.
The problem with this market hardly needs emphasizing.
To start with, market values of individual shares are
not a reflection of the true worth of the companies
involved, but the state of demand for those shares relative
to their supply. Demand, at the margin, was clearly
being driven by foreign institutional investors in search
of diverse portfolios, who then numbered more than 500
in the country. During the first 18 trading sessions
in February 2000, the FIIs invested more than Rs. 2660
crore, which was more than half the annual average investment
that occurred during the years since 1993, when FII
investment in India's stock markets was first allowed.
The inflow during those days was in fact higher than
the inflow during the last six months of 1999. Clearly,
there had then been a sudden surge of interest in India.
With hindsight, three factors appear to explain this
trend. First, a speculative boom in IT stocks in general
and internet stocks in particular in American markets.
Second, the fact that among emerging markets India was
a country with a growing IT presence, in US markets,
strengthened by strategic alliances with leading US
firms. Finally, the fact that those Indian firms that
had gone in for a NASDAQ listing in American markets
were at that time performing quite well, encouraging
other Indian firms in the IT and entertainment sectors
to contemplate a similar strategy.
These factors had ensured that the speculative fever
in IT and related stocks in American markets had spilt
over into the Indian market. For example, the bull run
in Wipro shares came in the wake of two major strategic
alliances it had forged with Microsoft, the software
giant, and Symbian, the combination of leading players
that were targeting the emerging market for wireless
devices that link to the internet.
While demand for Indian equity in selected sectors was
spurred by these factors, the supply of such shares
was (and is) limited for two reasons: first, internationally
acceptable players were still small in number, even
if increasing over time; and, second, the number of
shares from such enterprises that was available in the
market was limited. As mentioned earlier, only 25 per
cent of Wipro shares were with the "public"
as opposed to the promoter, and most of those holding
such shares were unlikely to be ready to part with them
in the course of a boom. The net result is that the
demand-supply balance at the margin is heavily weighted
in favour of sellers, resulting in astronomical price
increases in short periods of time. This is true of
other companies as well. Needless to say, if many promoters
chose to exploit the situation by off-loading a significant
chunk of their holding, the demand-supply balance for
shares of individual companies could change substantially,
resulting in a fall in prices that is as dramatic as
the previous rise.
Despite this dependence of share prices on the limited
supply resulting from a high holding by the promoter
and their associates, the market capitalisation index
applies the price at the margin to value the stock of
the company. This results in a dramatic surge in the
"market value" of the company along with the
price. Not surprisingly, a few firms and sectors accounted
for the early-2000 surge in market capitalisation. By
mid-February, Wipro alone accounted for 15 per cent
of market capitalisation in the BSE and the combined
market value of around 150 software companies accounted
for 32 per cent. It must be remembered that at the beginning
of the 1990s, these companies hardly featured in the
BSE.
In short, India's new found wealth was like a pyramid
of cards built by a bunch of flighty investors. Small
money by world standards was rushing into a few sectors,
honing in on a few companies which had a small number
of shares on trade. This pushed up prices at the margin
to create an illusion of wealth, because the commodity
producing sectors, especially agriculture, were languishing.
But at the top, things appeared as if they could not
have been better.
However, that speculative fever gave way to a sharp
fall in stock prices when investors (foreign and domestic)
realized that there was nothing which warranted a share
to trade at 750 times the annualised per share earnings
of a company, as it did in the case of Wipro. Compared
to its February 2000 peak of close to Rs. 9000 per share,
the Wipro stock closed at Rs. 2282 on October 20, 2000,
which was the day after it made its debut at the New
York Stock Exchange.
The focus on and dominance of a few shares was explained
by another feature of the IT industry. This was the
overall concentration of exports and sales in a few
domestic firms in the industry. In periods when IT stock
prices rose, it is the stock of these companies that
soared, converting their promoters into billionaires
even for a short period of time. However, the performance
of the few other firms that were listed in the stock
market was lacklustre. This difference in stock price
performance between the leaders and followers has persisted
to this day.
What has changed however is the average relative performance
of the sector. Investors having burnt their fingers
during India’s version of the dot-com boom and bust
at the turn of the millennium have since held back investments
in this area. As a result, the maximum level that the
IT index touched during the post 2003 period (on 19
February 2007) was 65 per cent below its 2000 peak.
The net result is that the IT sector has been losing
its importance in the market as a whole. The ratio of
the market capitalization of three leading firms for
which data is available for a long period to the BSE’s
aggregate market capitalization has fallen from more
than 25 per cent in March 2000 to less than 4 per cent
by October 2007 (Chart 3).
Chart
3 >>
This growing divergence between the market in general
and the IT firms within it indicate that after the experience
of 1999-2000, investors moved on to other shares, so
that the boom since 2004 has been focused on other industries
and sectors. But the investors are the same or similar,
as evident from the fact that FIIs have played a major
role in the post-2003 boom as well. This raises a question
that looks for an answer: now that the post-2003 boom
in stock markets is reversing itself, where would these
investors turn? Commodities and commodity futures may
be one answer. If so, those calling for increased regulation
of these markets may have a case, because the effects
of speculation there adversely affects those who have
gained little, if anything, from India’s post 1990 growth.
|
|
Print
this Page |
|
|
|
|