Friday 10 August 2007

Jack Kilby--A Tribute

Business India, November 13-26, 2000

When the chips are up

Jack Kilby, inventor of the integrated circuit, gets his due with the Physics Nobel Prize 2000 - after 42 years

Shivanand Kanavi

Nearly 42 years after the invention of the integrated circuit (IC) the Swedish Royal Academy of Sciences announced on 10th October 2000, the award of Nobel Prize to Jack Kilby, the inventor. Naturally, this calls for comment.

It is surprising that the Nobel Com­mittee took such a long time to hon­our an invention, which has changed the way we live in the last 40 years and has directly led to a $231 billion industry. The committee is still to honour Claude Shannon whose semi­nal contribution to information the­ory is at the heart of all modern communication and is over 50 years old. Similarly, John von Neumann, who laid the foundation of computer science with the theory of finite automata, 50 years back, and who led the team that built the world's first electronic computer ENIAC, at the Institute of Advanced Study, Prince ton, did not receive the Nobel. How­ever, inventors of semiconductor devices like the transistor and Joseph­son junction have been honoured. Does the committee have a blind spot? Or should the engineers be left out of the highest scroll of honour because of a very narrow-minded reading of Alfred Nobel's charter? Even if the rest of the technology world discussed these issues, Jack Kilby seemed far away from it all, when Business India interviewed him 48 hours after the announcement, in Dallas, Texas.

Kilby's brevity and humility are astounding. Texas Intsruments (TI), the company where Kilby worked when he invented the IC, made full use of the propaganda value of the prize and rival Intel's executives com­mented - off the record - that the prize should have been shared between Jack Kilby and the late Dr Robert Noyce (one of the founders of Intel). But Kilby, 77, could not be drawn into any controversy.

On 12th October, Business India was privy to a very private celebration at the famous Kilby Centre at Dallas, Texas. The $ 150-million cutting edge R&D centre, the pride of TI , hosts state­of-the-art, 0.1 micron chip fabrica­tion facilities. In the cafeteria of Kilby centre, about 300 TI engineers were gathered to honour an icon. The atmosphere was bordering on the rev­erential, as aging Jack Kilby slowly walked into the hall escorted by K. Bala, senior VP at TI. While Bala, a chip veteran for the last 31 years at TI, and the senior-most executive of Indian origin in TI, gave a two-minute speech congratulating Jack on behalf of TI and the staff of the Kilby Centre, the man at the centre was even more brief. True to the culture of micro­miniaturisation, Kilby only said two words: "Thank you".

Besides dozens of other honours, Kilby received the National Medal of Science, in 1970, at the White House. In 1982, he was inducted into the National Inventors Hall of Fame, tak­ing his place alongside Henry Ford, Thomas Edison, and the Wright Brothers in the annals of American innovation. However, there is noth­ing like being bestowed with the Nobel. When asked why it was that the inventors of transistor were given the Nobel several years ago, he com­mented candidly that their contribu­tion was more basic to semiconductor physics, than his.

Kilby is an institution at TI, even though he stopped doing active work in the company more than 20 years back. "We have a regular Friday lunch with him every week and find interact­ing with him very useful even now,” says Bala. The unassuming Kilby is a typical engineer, who wants to solve problems. In his own words, his inter­est in electronics was kindled when he was a kid growing up in Kansas. “MY dad was running a small power com­pany scattered across the western part of Kansas. They had a big ice storm that took down all the telephones and many of the power lines, so he began to work with amateur radio operators to provide some communications and that was the beginning of my interest in electronics," he recalls.

When he joined TI in the summer of 1958, the company was working on a defence project to miniaturise elec­tronics. Jack, however, had different ideas. Fortunately, he got the chance to check out his ideas, for, soon after he joined, most of the team went on a vacation for which he, being the junior-most member of the team, was not eligible.

Kilby spent his time searching for an alternative to the official model. "I realised that semiconductors were all that were really required - that resistors and capacitors (passive devices), in par­ticular, could be made from the same material as the active devices (transis­tors). I also realised that, since all of the components could be made of a single material, they could also be made in situ and interconnected to form a com­plete circuit," Kilby wrote in a 1976 article titled Invention of the IC.

"I then quickly sketched a pro­posed design for a flip-flop using these components. Resistors were provided by bulk effect in the silicon, and capacitors by p-n junctions. My col­leagues were sceptical and asked for some proof that circuits made entirely of semiconductors would work. I therefore built up a circuit using dis­crete silicon elements. Packaged grown-junction transistors were used. Resistors were formed by cutting small bars of silicon and etching to value. Capacitors were cut from dif­fused silicon power transistor wafers, metallised on both sides. This unit was assembled and demonstrated on 28 August 1958," he elaborated.

By September, Kilby was ready to demonstrate a working integrated cir­cuit built on a piece of semiconductor material. Several executives, includ­ing former TI chairman Mark Shep­herd, gathered for the event on 12 September 1958. What they saw was a sliver of germanium, with protruding wires, glued to a glass slide. It was a rough device, but when Kilby pressed the switch, an unending sine curve undulated across the oscilloscope screen. His invention worked - he had solved the problem.

But at the time, did Kilby realise its significance? "I thought it would be important for electronics as we knew it then, but that was a much simpler business. Electronics was mostly radio and television and the first comput­ers. What we did not appreciate was how lower costs would expand the field of electronics beyond imagina­tion. It still surprises me today. The real story has been in the cost reduc­tion which has been much greater than anyone could have anticipated. And it has tremendously broadened the field of electronics. In 1958, a sin­gle silicon transistor that was not very good, sold for about $10. Today, $10 will buy over 20 million transistors, an equal number of passive compo­nents, and all of the interconnections to make them a useful memory chip. So, the cost decrease has been factors of millions to one. And I'm sure that no one anticipated that".

The main contribution to lowering costs and increasing transistor densi­ties in ICS has, of course, come from the innovation that Robert Noyce was responsible for, while working at Fairchild Semiconductors, before he went on to found Intel. Noyce devel­oped the planar technology of deposit­ing layer upon layer of semiconductors with different doping elements to cre­ate a really compact IC in 1959. Clearly, both Kilby and Noyce are responsible for the IC revolution. Kilby graciously admits as much. The first Ie made by Kilby had a single transistor, 1.5 mm x 10.5 mm in size. Today, such chips carry millions of transistors. The reso­lution of etching has already reached 0.1 micron (micron = thousandth of a milli metre).

The inventions of Kilby and Noyce have led to two giants now straddling the IC empire: Texas Instruments, the king of digital signal processors that go into every cellphone and communica­tion device, and Intel which is practi­cally inside every PC. Unlike most inventors, Kilby today has the rare advantage of seeing his work create enormous wealth and in fact changing human life. In his own life time.

Reliance Story

Business India, July 14-27, 1997

Playing to win

With the commissioning of the largest multi-feed cracker in the world at Hazira, Reliance catapults itself into the world of petrochemical giants

Shivanand Kanavi

Scene 1: The place: Oriental Hotel in Sin­gapore. The event: A discreet meeting of major polymer producers from Korea, Indonesia, Thailand, Malaysia, Singa­pore, Japan and Saudi Arabia, with the Americans joining later for an 'informal' dinner. The host: A senior executive of an Indian company. The agenda: Problems of the global plastics industry, falling poly­mer prices and dumping in each others' markets.

Such a get-together is taking place for the first time. Once the pleasantries are over, the Indian executive softly informs the others that if they continue to dump polyethylene into India then he will pay them back with interest in their home mar­kets. After all, the executive says, all it takes is a small, well-directed consignment to spoil a market. The words are heeded. The dumping stops soon after.

Scene 2: Another time and another place. Some European producers of linear alkyl benzene (LAB) are similarly cautioned by this company against dumping into India. This time, too, the words are heeded

The Company: Reliance

Reliance provokes extreme reactions in India. Either you are pro-Reliance or anti-Reliance. There is nothing in between. To many, God is a poor second cousin to Dhirubhai Ambani, the chairman of Reliance Industries Ltd (RIL). To oth­ers, the Ambanis typify the seamy side of big business: fixing import quotas, pre-empting licences and switching share certificates. Amidst such extremes, public drama, recrimina­tion and bitter controversies, RIL has built up huge capacities at extremely competitive costs and become a force to reckon with in petrochemicals.

This is a story of dominance. In the licence regime, RIL used every trick in the book to control competition and carve out niches in whichever market they operated. With the gradual open­ing up since the late 1980s, Reliance changed gear. Despite their formida­ble political and bureaucratic clout and their profound understanding of power, Dhirubhai and his sons, Anil and Mukesh Ambani, realised that the new globalised game would be one of scale, competitiveness, technological prowess and market dominance. Hav ing created one of India's greatest syn­thetic textile brands - Vimal - the troika decided that real power came with backward integration and huge, globally comparable scales of opera­tion. Once the strategy was decided, RiL executed it with a vengeance, even at the cost of short- and medium-term shareholder value.

With his recent dramatic announcement of the 1: 1 bonus share issue after a hiatus of 14 years, Dhirub­hai is again back to being the God of the shareholders. Divinity aside, what Business India found was a remarkable story of the emergence of a highly competitive mega-corporation with skills and vision required to compete in a globalising economy. Well on its way to hitting $8-10 billion by the year 2000, the Reliance group is possi­bly the only real global giant in India.

From cloth to cracking
RIL emerged in the early 1970s as a pioneer in branded textiles. In the early years, when RIL'S Naroda mill was in its infancy, Dhirubhai and his cousin, the late Rasikbhai Meswani, built the brand with cloth produced by thousands of powerloom weavers in Surat. Excellent production and finishing supported by a sustained media campaign that lasted for over two decades made Vimal a household word. As if that were not enough, RIL became the company that developed the equity cult. With public offers of shares and partially convertible debentures, RIL went straight to small investors to raise huge sums of money long before the phrase 'primary market' entered the Nariman Point lexicon.

In 1982, RIL honed in on backward integration. Given the polyester con­tent in Vimal fabrics, the obvious choice was to manufacture polyester staple fibre (PSF) and polyester fila­ment yarn (PFY). The PFY plant was launched in Patalganga (Maharash­tra) in 1982, which was followed by PSF capacity in 1986. The next step was to move backward into purified terephthalic acid (PTA) in 1986 and paraxylene (input for PTA) in 1988. The market for detergent input was captured by setting up a capacity to manufacture linear alkyl benzene (LAB) in 1987. Then, in 1991, the Reliance group went in for the most ambitious project of them all at Hazira. Besides housing the largest single multi-feed ethylene cracker in the world, Hazira is a complex of a dozen global-sized downstream plants producing polyethylene, polypropylene, polyvinyl chloride (PVC), PFY, PSF, polyester terapthalate (PET) and intermediates like vinyl chloride monomer (VCM), MEG and PTA. The cracker plant was commis­sioned in March this year and, with the exception of the PET and the sec ond PTA plant, the Hazira project has been completed.

"There has never been in the his­tory of chemical industry in India something as big as the Hazira com­plex.," says internationally recog­nised chemical engineer M.M. Sharma, whose effusiveness may be partly because Mukesh Ambani was his student. But, the fact remains that the combined annual cracking capac­ity of IPCL (at Baroda and Nagothane), Nocil and the smaller crackers put together is less than Hazira's capacity. An unabashed Reliance fan, Sharma says, "Knowing their capacity to debottleneck, put balancing equip­ment and expand, I will be disap­pointed if the Hazira plant does not produce a million tonnes of ethylene within two years."

The six-tenth rule
The technological key to RIL'S expan­sion in petrochemicals is increasing returns to scale. Continuous process chemical industries are characterised by cylindrical pipes and reactors. There is a simple mathematical prop­erty that the area of the cross-section of a cylinder increases as a square of the radius. This yields the famous 'six-­tenth rule' that is beloved of chemical engineers: if capacity rises from 300,000 tonnes per annum (tpa) to 750,000 tpa (i.e. 2.5 times), then the cost of production will only increase by a factor of 1.73. Of course, the mathematics makes even more sense if one adds the purely commer­cial advantage of negotiating bargain­ basement prices in bulk purchase of plant equipment and raw materials ­not to mention the ability to use giant scales to pre-empt competition.

While the advantages of econo­mies of scale are well known, there are serious obstacles in putting up glob­ally sized chemical plants in India, such as the lack of infrastructure to handle raw materials, inadequate port and jetty facilities, erratic power and water supply and non-availability of huge funds at internationally compa­rable costs. What sets Reliance apart is the manner in which it has levelled this uneven playing field by meticu­lous project planning, excellent sys­tem engineering and swift project implementation backed up by the trading savvy of three of the sharpest entrepreneurial minds in the country. To overcome infrastructure prob­lems, RIL has gone ahead to create self ­sufficient islands of in-house facilities. For example, it has built its own jetties in Hazira and a single buoy mooring 5 km off the coast for large tankers to transfer liquids directly into the stor­age space (called tank farms). All RIL sites are self-sufficient in power. The infrastructure that has put RIL on the world map is the ethylene terminal at Hazira. Since the cracker plant was to be commissioned at the very end, RIL needed large-scale imports of ethyl­ene for their downstream pvc and PET plants. So they built a cryogenic ter­minal for transferring ethylene at ­-138°C in deep seas. "That was the first time it was done in the world, " recalls Anil Ambani. "Everybody told us eth­ylene transfer is unsafe and definitely not possible in a country like India. But once we proved that we can do it, and do it safely, others in the world are copying it."

Sweating it out
“A major strength of RIL is the way they integrate their projects," says Pradeep Shah of Indocean Venture and a long-time Reliance watcher. The importance of integrating back­wards is a lesson that RIL learnt from the Patalganga project. PTA and DMT require paraxylene as feedstock, which had to be imported. Since India had hardly any infrastructure to han­dle imported chemicals, the Ambanis had no hesitation in going for a paraxylene plant with technology from the world leader, Universal Oil Products (UOP). Their next-door neighbour, Bombay Dyeing, which had put up a dimethyl terapthalate (DMT) plant, suffered because it depended on imported paraxylene.

The paraxylene plant required naphtha as feedstock. Instead of rely­ing on tanker movement by road from Mumbai to Patalganga, RIL quickly laid a pipeline to Bharat Petro­leum's refinery. When they discov­ered that the hydrogen produced in the catalytic cracking of naphtha was being flared away, RIL diverted it to the PTA plant where hydrogen is needed. Similarly, instead of wasting the nitrogen by-product in the PTA plant, RIL used it as a conveyor gas. By­products of naphtha cracking are used in gas turbines, and kerosene, from which paraffins are extracted for mak­ing LAB, are similarly burnt. Any extra heat in a process is not wasted, but used to pre-heat various feedstock.

But system and backward integra­tion are not all. According to Mukesh Ambani, one of the main reasons RIL's 1996-97 performance was better than expected by analysts is the use of 'sweat technology,' or making the plants produce more through clever chemical engineering. Teams work relentlessly to milk as much as possi­ble from the plants. For example, the engineers at Patalganga discovered that adding another compressor and supplying more air to the reactor where paraxylene gets oxidised in the PTA plant would enhance the output greatly. Immediately, the cost-benefit ratio was worked out and an expen­sive 30-tonne compressor added. Today, the engineers are not fully sat­isfied. Up to now, only 23 tonnes of air are being used. So a team is work­ing on how t0 use the remaining seven tonnes.

Clearly 'sweating' pays off. A case in point is PVC production. IPCL uses 420 cubic metres of total reactor vol­ume to produce 50,000 tpa of PVC, while Finolex uses 500 cubic metres to produce 110,000 tpa. In contrast, RIL'S first pvc plant at Hazira utilises 560 cubic metres to produce 180,000 tpa. "When our people went recently to Geon, the company which supplied the technology for our I've plant, they were told Reliance has nothing more to learn in PVC technology," says K. Ramamurthy, president of RIL's polymer division.

The Vanvas is over

RIL's triumphal return to its shareholders

People had written him off as an old man who had lost his magic touch. On 26 June, 1997, in the middle of RIL's annual general meeting, Dhirubhai Ambani announced an interruption with a board meeting to discuss bonus shares. The mar­ket began to oscillate wildly. Then Dhirub­hai came back and asked the shareholders, "What do you want?" The roar of "One is to one!" reverberated across Birla Matu­shree Hall. He grinned, paused, and said, "You have it." The market went gaga as Lord Dhirubhai returned to his throne.

With the bonus issue after 14 years, the Ambanis have outfoxed equity analysts once again. "We need to be good at psy­choanalysis to predict their results," says an analyst jocularly. According to Jal Irani of Jardine Fleming, "Petrochem analysts have difficulty in quantifying the effect of economies of scale on manufacturing costs and hence the bottomline." Read that as a confession, "We don't know."

But, corporate watchers are happy to eat crow, delighted as they are to see one of India's fastest growing companies churning out a better-than-expected per­formance. Analysts predicted net profits ranging from Rs900-1, 100 crore. Instead, RIL overcame depressed petrochemical prices to post a net profit of over Rs1 ,300 crore in 1996-97, which was 1.3 per cent higher than the previous year. It managed to buck the industry trend by pushing vol­umes. The turnover for 1996-97 (Rs9,31 0 crore, of which Rs290 crore is other income) is 12 per cent higher than the pre­vious year.

The view of analysts is that despite the controversies that have dogged Reliance, "it has
created assets for all to see." The asset growth is quite phenomenal. Between 31 March, 1992 and March, 1997, RIL's total assets have increased at? compound annual rate of almost 28 per cent for the last five years and now stand at Rs19,536 crore. Last year alone it increased by 30 per cent.

This growth has been financed through higher leveraging. At a time when India's equity market was in the bear phase, the Ambanis were the first major to tap the overseas debt market with long ­term debt, including the 100-year Yankee bond. As of March, 1997, about 62 per cent of the total debt is foreign currency denominated. However, such leveraging carries risk. RIL's overall debt/equity ratio has risen from 0.49:1 in 1995-96 to 0.83:1 in 1996-97. Consequently, the interest bill has increased by over 54 per cent from Rs110 to Rs1 70 crore. This, plus a 22 per cent jump in depreciation (to Rs 410 crore) and Rs45 crore on account of the minimum alternate tax (MAT) has con­tained RIL'S post-tax profit at Rs1 ,323 crore - up by only 1 .3 per cent over that of last year. The rise in depreciation was because RIL responded to MAT by changing the method of calculation from straight-line to written-down value.

Have the results been dressed up? Some analysts have asked how a 35 per cent increase in volume can be achieved with a less than 12 per cent rise in manu­facturing expenditure. Granted that raw material costs have come down, but labour, power and other costs, plus infla­tion, should have pushed manufacturing costs higher. The Ambanis attribute this to ruthless operational efficiencies. "We make our plants sweat and try to squeeze the maximum out of them," says Mukesh.

The fact is that RIL has successfully with­stood the pressures of 1996-97 - one of the most difficult and recessionary years for the petrochem industry. Today, the company has world-class plants, inte­grated facilities and state-of-the-art tech­nology. RIL also has one of the lowest employee costs (2.7 per cent of sales), although the capital cost of newly set-up plants would be higher when compared to depreciated facilities of major global com­petitors. While RIL is diversifying into oil, gas, power and telecom, its mainstay in next three to four years will be petro­chemicals and polyester.

According to some analysts, the cur­rent year will see RPL make full use of its capacity expansion and total integration from crude oil and cracking facilities to downstream fibres and cloth to post a phenomenal growth in sales and earnings. Mahesh Talreja of Ventura Securities places the estimated growth in earning for 1997-98 at 56 per cent and "an EPS of Rs45 on its pre-bonus issue equity." Rupa Bose of Banque Paribas differs. In her report, she predicts a 20 per cent increase in PAT to the region of Rs1 ,640 crore and a more modest 23-24 per cent growth in EPS. Jardine Fleming's latest PAT projec­tions are similar: Rs1 ,611 crore.

Even if the conservatives are right, it would be an impressive performance. But you can never tell with Dhirubhai. The old fox could stun the market yet again next year. If he does, the analysts will be seen at the Zodiac Grill ecstatically eating Crow a la Languedoc!

Calculated risks

"We can deal with the risks in petro­chemicals like the process design risk, feedstock risk, operational risk or the market risk as well as anyone else in the world," asserts Mukesh Ambani. But their record shows that it is no empty bravado.

The first attempt at risk-taking was at Patalganga, when RIL took its chances by putting up capacities that were bigger than the entire Indian market for PFY and PSF. It paid off. India's market for synthetic and blended cloth grew by leaps and bounds in the 1980s and early 1990s despite adverse excise duties vis-a-vis cotton. Soon afterwards, the decision to bet on the new technology from ICI of making PTA, a feedstock for poly­ester, was taken despite the fact that, at that time, DMT was the proven raw material. The risk was taken because the PTA route did not need an inter­mediate extraction plant and, hence, translated to a lower capital cost. Again, the risk paid off. Today the PTA-vs-DMT controversy has been globally settled in favour of PTA.

Similarly, when RIL was putting up its first pvc plant in Hazira in 1990­91, the rated output of 180,000 tpa was larger than the total Indian demand. Today, RIL has expanded their capacity in the same plant to 275,000 tpa, and the market has not only absorbed this output but that of others like Finolex and IPCL as well. And Reliance isn't satisfied. Rama­murthy of the polymer division is planning to increase the Hazira capac­ity to 310,000 tpa. As a Reliance watcher and family insider pointed out, "The hallmark of the father and sons is that after meticulous research they rely on their sharp trading instinct. It is this instinct that makes them incessantly bet on growth and virtually create the growth by generat­ing massive economies of scale."

The Jamnagar experiment

Reliance's emphasis on system inte­gration, which evolved at Patalganga, has become a way of life in project planning. While the Hazira complex has numerous examples of such inte­gration, RIL'S Jamnagar refinery might become a textbook case. When asked why the entire technology package for the ]amnagar project was awarded to UOP, Mukesh Ambani pointed out, "Large benefits are going to come from clever feedstock and heat inte­gration. That can only happen when the right hand knows what the left hand is doing."

All components of the output at Jamnagar are either end products or by-products that are to be used some­where as a feedstock for high-value petrochemicals or as fuel for generat­ing process heat. In fact, petrochemi­cal experts estimate that, when the entire system stabilises, no more than 0.2 per cent of the refinery through­put will be wasted. If that happens, it will be a world record, because the international wastage figure for refineries hovers around 5 per cent.

RIL's system planning often takes place well before the project is imple­mented. For instance, when BHEL engineers were asked four years ago by Reliance to experiment with petro­leum coke as boiler fuel, they were naturally puzzled. It is now evident that even in 1993-94, the RIL team was looking at ways to use refinery residue - six years before the complex was due for completion. The experiment worked. Now, the gameplan is to con­vert refinery residue into petroleum coke to be used for the in-house power project to generate 1,000 MW of power.

When phase one is completed in 1999, the Jamnagar refinery project will be truly mind-boggling. It will not only supply 3 million tonnes of naph­tha to Hazira but also enough propene as feedstock for another 400,OOO-tonne polypropylene plant and a 1.2-million tonne paraxylene complex in Jamna­gar. Petrochemical engineers expect that the second phase at Jamnagar (starting from 2001) will produce another 1 million tonnes of ethylene as well as high-value petrochemicals for use in drugs, dyes and paints, unleaded petrol, food additives and polystyrene foam. In fact, a senior technologist at uop said that RIL is actually driving refining technologies in new direc­tions - away from traditional refining towards high-value petrochemicals.

Jack be nimble ...

"Without flexibility in operations one cannot quickly respond to market conditions," says Mukesh Ambani. Said by many, done by few. In Reliance's case, flexibility seems to be the holy word - be it in engineering, systems design or financial deal-mak­ing. Consider just the production side. The polyethylene plant at Hazira has a swing reactor that can produce, with hardly any change, either high-density polyethylene (HDPE) or linear low-density polyeth­ylene (LLDPE) according to market requirements. The cracker at Hazira is designed to use gas, liquefied natural gas, naphtha or gas oil as feedstock. "Since feedstock prices cannot be pre­dicted, RIL has the flexibility to use whichever combination that increases our operating margins," explains Mukesh Ambani. "Consider­ing the lifecycle cost of the cracker, the extra investment on a multi-feed cracker is marginal."

This agility translates to greater value addition. Earlier, RIL was spend­ing $1,300 on ethylene to produce about $2,800 of products. Today, the cracker uses no more than $600 worth of naphtha to produce the same end value. "We hope to make a profit of Rs2,200-3,000 crore every year from Hazira alone," exults Ramamurthy. According to him, even if global petro­chemical prices fall by 10-15 per cent, the high value addition at Hazira will sufficiently insulate RIL from such downturns. "On top of it, our market­ing. costs are extremely low," says Ramamurthy. "We spend $3 per tonne for marketing $800 of plastic. Our total distribution cost of plastics to any point in India is $26 a tonne. Let anyone who wants to enter the Indian market beat that!"

New horizons
RIL has never been shy of seizing a business opportunity: Here is a peek at some of the things they're getting into
Reliance Power is a separate company wholly-owned by RIL. A power purchase agreement (PPA) has been signed for the 41 0 MW Patalganga project and is now awaiting techno-economic clearance for financial closure. The Bhawana 423 MW project is waiting for PPA to be signed. Reliance is also bidding for a 2,000 MW super thermal pit head project in Bihar and a 800 MW hydel project in Himachal and 1,200 MW hydel project in Sikkim.

Reliance Telecom, another separate com pany with 10 per cent participation from Nynex, has won cellular licences for seven circles: Madhya Pradesh, Orissa, West Bengal, Bihar,Sikkim and the seven sisters of the northeast. RIL was the first to sign the agreement with DoT to pay the required fees and is in fact well on its way to starting cellular service in 22 cities.

Oil & Gas
Oil & Gas is a division of RIL. The ONGC- Enron-RIL (40:30:30) joint venture is pump­ing out over 12,000 barrels of oil daily from Panna and Mukta. These fields were termed marginal and uneconomical by ONGC. RIL found that with their project management expertise the cost of recov­ery could be lowered 20-30 per cent and, with Enron agreeing to a lower internal rate of return, the project became viable.

They have also hit gas at Tapti. One million cubic metres of gas will soon flow daily to Hazira. Since there are problems in sharing the existing pipeline with ONGC, RIL is trying to convince its partners to go in for a separate gas pipeline. Another promising deal is being nego­tiated in Iraq between the Iraqi govern­ment and RIL-ONGC videsh. If it comes through, this field should yield in one day (about 200,000 barrels) the entire production from Bombay High.

Master builders
While RIL'S low capital and opera­tional costs are recognised globally, another factor which makes them competitive is the speed at which they erect plants. All the Hazira plants were erected in record times, with the giant cracker coming up in only 29 months. All indicators point to Jamnagar pro­ceeding at an even more frenetic pace. According to Sharma, "The main rea­son is that the Ambanis sit on every­body's head, be they contractors, sub-contractors, or employees, and put in long hours themselves. Do you know that there were a couple of CMD's of IPCL who never spent a night at Nagothane when it was com­ing up, even though it was their biggest investment at that time?" Hetal Meswani, a cousin of Mukesh and Anil and an executive director of RIL, was literally living in Hazira dur­ing construction and has now moved to Jamnagar. Engineers recall how Mukesh Ambani would spend days on end at Patalganga when it was coming up.

Today there are daily video confer­ences between Mumbai, Jamnagar and the London office of Bechtel, who are doing the engineering work for the refinery project. There is a team of RIL engineers sitting in the London office of Bechtel, while a team of Bechtel engineers is at Jamnagar. New tech­nologies are being adopted to speed up construction. "We found that 40 per cent of the time in erecting piping is taken up by welding the joints," says S.C. Malhotra, technical executive director of Reliance Industrial Infra­structure, a Reliance group company. "So we decided to spend a few crore and automate the whole thing by going for a giant induction pipe-bend­ing machine that can bend even a 2-ft diameter steel pipe into any angle."

Blurring boundaries
"Do you think a modern Rs10,000 ­crore business can be run by two brothers?" asks an executive rhetori­cally. From foremen, site supervisors and middle-level plant managers to vice-presidents and presidents, every one in RIL is given considerable auton­omy. According to Pradip Shah, that is one reason why Reliance has such a strong core group of highly talented professionals in marketing, finance, operations, projects, construction, telecom, power, oil and gas, fibre, intermediates, polymers and textiles. It is a peculiar but successful amalgam - the family makes the key decisions, which are then meticulously executed by professionals.

"In the best companies of the world the boundaries between technology, finance, and marketing have collapsed," points out Mukesh Ambani. Just as the siblings don't have any rigid compartmentalisation of responsibilities, so too do many of the top executives of RIL. "I was coming from a conference in Singa­pore and I heard about a good trading opportunity in Jakarta," says C.S. Gokhale, who runs the fibre intermediates division. "So, I rang up Mukesh and he not only agreecd with me but asked me to come via Jakarta and clinch the deal. There are no bar­riers iuch as this is a purchase func­tion and somebody else has to do it. After all, when I joined RIL, I was told that I will be an owner-manager, and I have no doubts on that." Hemant Desai, vice-president in charge of polymer plants at Hazira, also empha­sises the importance of being a well­rounded manager. "I used to be in fibre marketing; today I am in poly­mers; tomorrow if I have to go to Jam­nagar and talk to the sarpanch of Mothikavdi [the village next to RIL'S site] I will do that." The motto is if it benefits the company, go ahead and just do it.

It is that motto which has driven RIL relentlessly. In the process, it has made enemies, occasionally bent laws and liberally buttered bread on both sides. But even Reliance's worst detractor cannot ignore four simple facts: it has created a massive asset base in the fastest possible time; it has produced successful plants on truly global scales; its costs are internation­ally competitive in the truest sense of the term; and that has given it consis­tent long-term value to 2.6 million shareholders. No company can beat that. Not yet, anyway.

Sibling speak
Excerpts from interviews

We will benchmark with the top 100 companies in the world

• We have always believed that we shouldn't get into anything because it was in fashion or everybody else was doing it. We must first understand it thoroughly. Then we back it up with dedicated leader­ship, which is not restricted to my father or myself and my brother.

• Our projections for the group in 1999-­2000 are assets worth Rs30,000-35,000 crore and a turnover between Rs40,000­-50,000 crore.

• If the administered pricing mechanism for petroleum is lifted and crude is decanalised, then with the way we have . configured the Jamnagar refinery, Reliance will be globally competitive.

• Refineries of the future will be fully inte­grated with petrochemicals and power. We were the first in the world to think that way and the first to build such a refinery.

• After Jamnagar, we will grow upstream into oil and gas and energy. For that we might even have to go outside India, and have already bid for Iraq sites with ONGC videsh. The new exploration policy itself will open up opportunities in India. If there is any oil in India then Reliance will defi­nitely find it.

• Right now we are having discussions with IOC about fuel distribution from the Jamnagar refinery. There is merit in India's oil majors getting together to reduce distri­bution costs.

• We won't move into a gas cracker pro­ject in Assam until it is clear what feed­stock we are going to get. Only then can we employ our knowledge and add maxi­mum value to the project. In India, people are not used to up-front rigorous analysis. Instead, they prefer foundation stone-lay­ing ceremonies. Let us first be clear about what we will get. That is better than to assume something and have problems later.

• Reliance will get listed on Wall Street because we want to benchmark ourselves with the top 100 companies of the world.

We have the lowest capital expenditure costs globally

• Our capex costs are the lowest because we do not look at a plant as a black box. We break it into every possible component, and optimise what we can do in India and what we can do offshore. Our aggression also comes from scale. We believe in hav­ing rate contracts. Also we buy everything with cash. If the vendor has to provide finance, he jacks up the price by say 20 per cent. Similarly we give the site contractors assured work in exchange of which they are ready to take a knock on the margins.

• The skills we built in the chemical indus­try we are repeating in telecom. Our costs are far lower than any other telecom com­pany in India. Investors are applauding our telecom strategy - to pay the lowest pos­sible licence fee and aim for the highest geographical spread. This will give us the ability to innovate in marketing and tariffs.

• We are not going to have 30 telecom operators in India. There will be a shake­out. Nearly 20 licences, including HFCL's, are up for sale. If the rules of the game change, then Reliance will be in the front­line buying some of them out.

• What's the big deal about P/E ratios? In the 1980s and 1990s, Shell was quoting at two times earnings! Isn't Shell a great com­pany? In the lifecycle of every company, there are ups and downs. Look at IPCL. They have no merger, no private place­ment, no seeming controversy, are a strong petrochemical player, have scale, integration, depreciated assets, 51 per cent owned by the government, no fear of great dilution, all accounts are certified by any number of government agencies. Then why is it quoting at four times earnings and RIL is quoting at ten times earnings? There is no rationale.

• As India integrates with the global econ­omy, people will begin to understand the value of assets. For people who are inter­ested in looking at the long term, these are very exciting times.

Book Review-- Maths-S Ramanujan

Ramanujan: Letters and Commentary (History of Mathematics, Vol 9)-- by Bruce C. Berndt, Robert A. Rankin

It is lonely at the top

Shivanand Kanavi

"I beg to introduce myself to you as a clerk in the accounts department of the Port Trust Office at Madras .... ". Thus started the most famous letter in 20th century mathematics, written by Srinivasa Ramanujan to Godfrey I-tarold Hardy.

Was Ramanujan, afraid that proofs of his theorems may be stolen by the dons at Cambridge or was he afraid of being ridiculed for his intuitionistic methods? Was he really afraid that he will be "polluted" by going to England or was he pressured into refusing to go by other orthodox Brahmins, and so on are among the many questions that are raised by this latest collection of Ramnaujan letters, edited by two mathematicians, B.C.Berndt and R.A.Rankin.

Ramanujan's method of stating an important result and providing some 'hand waving' arguments as the basis of the result rather than give rigorous proofs was unique and mathematicians like MJ.M.Hill dismissed his genius. One of the reasons for the cryptic nature of his proofs was his notation which he very much kept to himself and the other was that many a time his "proof" will involve leaps that could not be bridged by rigorous mathematics. Nevertheless G.H.Hardy recognised the genius and endeavoured successfully to get him to come to Cambridge. In fact the collection contains a letter written by Bertrand Russel to one of his women friends wherein he mentions, "I found Hardy and Littlewood in a state of wild excitement, because they believe they have discovered a second Newton, a Hindu clerk in Madras".

But actually, Hardy's close associate J.E.Littlewood thought otherwise, in a letter he told Hardy that, he placed Ramanujan on par with Jacobi (important Hungarian mathematician, but not up to the stature of Newton). He also thought Ramanujan was not disclosing his proofs because he probably feared that English mathematicians might "steal" his results! Ramanujan expressed his pain on reading the same allusion in Hardy's reply to his first letter, but surprisingly did not disclose any more information regarding his methods. Instead he wrote, "I am already a half starving man. To preserve my brains I want food and this is now my first consideration. Any sympathetic letter from you will be helpful to me here to get a scholarship either from the University or the government", and added a few more results in his famous second letter. Considering his economic plight and the craving for professional peer recognition and even critique and considering the plundering of India in the hands of Britain his unexpressed fears are understandable.

A misconception cleared by the collection under review is regarding Ramanujan's orthodoxy that. prevented him from accepting the invitation to go to Cambridge. In a letter to Hardy, Ramanujan explained that when the offer was made by C.E.Mallet of India Office, London, it was too vague and besides it brought visions of having to appear for a civil service examination. This made him hesitate and mean while, his boss, an orthodox Brahmin, who was accompanying him, intervened and categorically told mallet that Ramanujan cannot go due to religious reasons. It is clear from the letters that Ramanujan was actually looking forward to meeting the best mathematical minds of Europe.

The letters also bring out the loneliness he suffered during his illness first in England and then in India. However being extremely conscientious and grateful for the fellowship he kept sending new results despite severe illness.(The illness which was never diagnosed properly has now been diagnosed as amoebiasis of the liver according to the authors). In fact Hardy's letter to J.J.Thompson (Nobel Laureate physicist and president of the Royal Society) shows that fearing the worst and aiming to give some incentive to Ramanujan to fight the illness, Hardy lobbied for the early award of the Fellowship of the Royal Society.

Though after his death his family members engaged in a sordid exhibition of greed and propensity to cash in on respect and sympathy for Ramanujan, he himself exhibited his generosity by asking his family to actually give some money for scholarships to poor students.
The collection is a valuable addition to the growing literature about Ramanujan's life and work. The editors have taken particular pains to add fairly detailed biographical notes on all the individuals that are mentioned in the letters, thereby giving a flavour of associates and friends of Ramnaujan. They have also given extensive mathematical notes on the history and evolution of each of Ramanujan's propositions as they appeared in his letters. On the whole it is an enjoyable reading for mathematicians as well as lay public interested in the Indian genius.