Banianomics, J-Curves and the Character of Indian Capitalism

Capitalism in India has its own flavour. It grew from a commercial tradition shaped over centuries by fragmented markets, limited formal credit, weak infrastructure and uneven institutions. From these conditions emerged a distinctive way of organising capital, family, risk and opportunity. I call it Banianomics.

Its instincts are straightforward: preserve capital, judge credit carefully, protect reputation, turn working capital quickly, control costs, limit downside and recycle surplus relentlessly. These habits developed most visibly among communities involved in trade, finance, brokerage, moneylending and the movement of goods across distance.

I use the colloquial term “Bania” here as a broad commercial idea rather than a narrow caste description. Variations of the same logic appeared among Marwaris, Gujaratis, Jains, Chettiars, Sindhis, Parsis and other merchant communities that became unusually good at making commerce work where formal institutions remained incomplete.[1]

Over generations, habits became capabilities. Merchants learned to operate without waiting for perfect institutions. Trusted networks stood in for weak contracts. Reputation lowered the cost of doing business. Family capital substituted for unreliable finance. Tight working-capital discipline reduced dependence on large balance sheets. Sourcing, distribution and execution became competitive advantages.

Indian capitalism became very good at operating under constraint.

Versions of this system can still be seen across the economy, from small-town merchants to some of India’s largest corporate houses. The scale has changed enormously, yet many of the instincts remain familiar. Capital is committed carefully. Cash matters. Execution matters. Technologies developed elsewhere are acquired, adapted and commercialised. Distribution becomes a weapon. Proven models are pushed into new markets. Capital moves quickly towards the next opportunity.

This is a powerful economic system. It can create large fortunes, durable companies and exceptional returns on capital. It can also carry a poor country surprisingly far.

Catch-up growth rewards precisely these capabilities. When technologies, machinery and business models already exist elsewhere, much of the economic task lies in acquisition and execution: find what works, finance it, localise it, manufacture or assemble it, distribute it and scale it. Banianomics is particularly well suited to this stage of development.

I think the same Banianomics that once helped India advance is now beginning to hold it back. The challenge changes as an economy approaches the technological frontier.

The next semiconductor architecture, drug platform, battery chemistry, operating system or industrial technology does not already exist somewhere waiting to be sourced and commercialised. Someone has to create it.

That process has a very different economic shape. It can require years of spending before there is a product and longer still before there is profit. Experiments fail. Scientists leave. Products arrive too early. Markets develop too slowly. Entire technical approaches prove wrong. The assets are often intangible, difficult to value and incapable of producing cash for years.

The financial profile is a J-curve: heavy expenditure first, uncertain returns much later.

This requires a different appetite for risk and a different institutional architecture around capital.

India therefore needs a parallel system of frontier capitalism: pools of capital, corporations, universities, laboratories and public institutions willing to finance long J-curves, repeated technical failure, uncertain markets and years of expenditure before returns appear.

The issue is larger than R&D spending. Commercial sophistication and technological depth are separate capabilities. A country can become highly proficient at operating the machinery of modern capitalism while remaining dependent on technologies created elsewhere.

A great deal of economic and strategic power sits upstream with those who own the architecture, intellectual property, platforms, standards and technological primitives around which everyone else must organise. Companies downstream may still become large and highly profitable. They nevertheless operate within technological systems whose economics and strategic constraints were often defined elsewhere.

This distinction matters more as India becomes richer.

Banianomics excels at turning existing possibilities into businesses. It rewards commercial judgment, capital efficiency, adaptation, distribution and execution. These capabilities remain essential. India would weaken itself by abandoning them.

The next stage requires adding the capacity to create new possibilities.

Frontier capitalism supplies the patient capital, institutional tolerance for failure and technical ambition required to develop technologies whose commercial value may take years to emerge. Banianomics supplies the discipline required to turn those technologies into scalable industries.

India will need both.

The stakes extend beyond innovation policy. A country dependent on technologies created elsewhere also inherits many of the economics, standards and strategic choices embedded in those technologies. Its firms may capture substantial profits while a disproportionate share of technological power remains upstream.

Escaping the middle-income trap and becoming a genuine third pillar of the global economy will therefore depend on India’s ability to combine two forms of capitalism: the commercial intelligence it has spent centuries refining and the willingness to finance long, uncertain journeys into technologies that do not yet exist.

Banianomics turns institutional scarcity into capital efficiency

Banianomics developed in markets where formal credit was limited, infrastructure was uneven and contract enforcement was expensive. Merchant communities supplied part of the missing market infrastructure themselves. They financed inventory, extended credit, judged counterparties through reputation, moved goods across fragmented geographies, carried information between markets and collected payment where banks and courts were weak or costly.

Economic value came from circulation. The merchant earned by keeping goods, information and money moving through a system full of friction.

Indian merchant communities reproduced many of these capabilities outside India. Thomas Sowell’s discussion of overseas Indian communities is useful here because migration allows us to observe commercial habits outside a single political setting. Gujaratis, Chettiars, Jains and other trading groups carried credit judgment, reputation systems, frugality, inventory discipline and kinship-based trust into East Africa, Southeast Asia, Britain and North America. Much of the productive asset was intangible: a body of commercial knowledge, relationships and routines that rarely appeared on a balance sheet.[2]

Mohnish Pabrai’s Patel motel example in The Dhandho Investor compresses this operating logic into a single small business. A family acquires an existing asset with visible demand, uses debt or seller financing to minimise scarce equity, lives on site, substitutes family labour for hired labour, controls maintenance and household spending, raises occupancy through a lower cost base, repays debt from operating cash and then redeploys the accumulated equity into the next property. The logic is simple and powerful: commit little equity, control the downside, keep costs tight, extract cash and replicate.[3]

The operating system has six recurring rules

Table 1. Six recurring rules of Banianomics. Author synthesis.

Frugality plays a direct role in capital formation. Lower household consumption increases retained surplus, reduces dependence on outside equity and allows the family to control more assets over time. Family and community networks reduce transaction costs by supplying labour, information, trust and succession. Capital compounds alongside capability. Supplier knowledge, lender relationships, operating routines and experienced managers make each subsequent business easier to establish.

This helps explain one of the most striking features of Banianomics: sectoral mobility. A merchant family can move from commodities into distribution, manufacturing, property, finance, infrastructure or telecommunications because its deepest capabilities travel well across industries. Opportunity selection, capital mobilisation, cost control, counterparty judgment, distribution and cash recycling remain useful even when the product changes.

Over generations, this logic can harden into the diversified business group. Trading profits fund distribution. Distribution reveals manufacturing opportunities. Manufacturing creates property and financial assets. Internal cash flows and accumulated reputation eventually create an internal capital market. The family moves across sectors while carrying the same capital-allocation instincts with it.

This operating system is especially powerful where demand is visible and the main commercial variables can be controlled. Trade, consumer products, real estate, financial intermediation, logistics, hospitality, infrastructure and established manufacturing all reward procurement discipline, pricing, relationships, working-capital control and execution. Banianomics can therefore take a capital-scarce economy a long way through commercialisation, localisation and scale.

Table 2. Selected Indian large-cap economic archetypes and characteristic sources of advantage. Source: company disclosures and companion benchmark workbook [8].

I have seen a similar logic repeatedly in my own work with Indian corporate houses, particularly in discussions around M&A and capability acquisition. The growth logic is often remarkably clear. Find something worth a dollar, acquire it for perhaps sixty cents, improve or scale it, and create a credible path to realise substantially more than the original investment. Another question appears again and again: can this opportunity double or triple the capital invested within three or four years?

That framing shapes the kinds of opportunities that receive attention. Capital is expected to acquire something identifiable, improve it and produce a visible path to value. Long J-curves, prolonged losses and years of uncertain capability-building fit less naturally within that framework because the asset being created may remain difficult to identify and harder still to value for a long period.

In my conversations with Western corporate counterparts, I have generally encountered a greater willingness to fund bets with distant commercial payoffs, uncertain technical outcomes and several years of negative accounting returns. The willingness varies enormously by company and sector, but the contrast in capital-allocation temperament is often noticeable.

These instincts eventually flow into the organisation itself. They influence what management teams consider an attractive opportunity, how investment committees define risk, which people get hired, how performance is measured and what sort of business architecture emerges.

A capital-allocation philosophy therefore becomes part of the operating culture. Repeated over decades, it shapes the organisation’s sense of what constitutes a sensible risk, an acceptable time horizon and a worthwhile use of capital. At that point, Banianomics is no longer simply a method of financing businesses. It becomes a way of seeing opportunity.

Frontier leadership requires a different capital architecture

Fernand Braudel’s hierarchy of capitalism provides a useful bridge from merchant economics to technological power. The commanding positions in an economic system tend to sit where control over capital, information, finance, technology, coordination and risk remains scarce. The largest surpluses accumulate around capabilities that are difficult to reproduce. As production, processing and distribution become standardised, competition rises and economic rents tend to fall.[4]

The same hierarchy appears inside modern value chains. LVMH controls brand and desire. Apple controls product architecture and its ecosystem. Nvidia controls semiconductor architecture and the CUDA software ecosystem. Each company can outsource or purchase large parts of its physical value chain while retaining control over the scarce asset around which suppliers, developers and customers organise.

This is the level at which technological power begins to matter.

Indian capitalism has developed deep capabilities in procurement, manufacturing, infrastructure, localisation, distribution, financing, implementation and regulatory navigation. These capabilities can generate exceptional returns. Frontier capitalism adds another task: financing assets whose commercial value remains uncertain for years and whose eventual payoff depends on proprietary science, architecture, software, standards or intellectual property.

The J-curve changes the required behaviour of capital

Frontier innovation often begins with irreversible commitments, negative cash flow, technical failure and uncertain payoffs. Capital must be committed before technical feasibility, product-market fit or a scalable market has been established. Progress may require several rounds of failure before a commercial system emerges.

Creative destruction makes the problem harder. A company may have to weaken an existing franchise, divert capital from a profitable business or build a technology that could eventually cannibalise its own products.[5]

Spend → fail → experiment → invent → own IP → scale → harvest rents

This process requires institutions capable of funding a long gap between scientific possibility and commercial dominance.

US venture capital institutionalises part of this logic through portfolios in which a small number of extreme winners can dominate total returns. Europe has historically financed parts of the same J-curve through public research institutions, state-funded science, defence programmes, industrial policy and patient corporate investment. The institutional structures differ across countries and sectors. The underlying economic requirement remains the same: someone has to fund the years in which technical capability is being created and commercial value is still uncertain.

Table 3. Two capital architectures: Banianomics and venture/IP capitalism. Author synthesis.

That distinction also changes how returns on capital should be read.

ROCE measures efficiency; reinvestment determines duration

ROCE = EBIT ÷ Capital Employed

or:

ROCE = Operating Margin × Capital Turnover

ROCE is especially useful for businesses whose advantage comes from making an accounting capital base work harder. Faster inventory turns, supplier credit, higher store productivity, greater factory utilisation, denser distribution and disciplined project execution can all lift returns.

An 8% operating margin combined with 5.0× capital turnover produces a 40% ROCE. That can be an outstanding business even when technological differentiation is limited.

The next question is what happens to the next unit of capital.

ROIC = NOPAT ÷ Invested Operating Capital

Economic Spread = ROIC − WACC

Growth = Reinvestment Rate × Incremental ROIC

For long-term value creation, the productivity of new capital matters more than the return reported on a mature asset base. A company earning a 30% current ROCE with little room to reinvest may have limited capacity to compound. Another company may earn 20% on capital and still create far greater value if it can redeploy large amounts of capital at similar returns for twenty years.[6]

Duration, reinvestment capacity and incremental ROIC therefore sit at the centre of the analysis.

This becomes even more important in frontier businesses because accounting statements capture only part of the capital being employed.

R&D is usually expensed even when it creates assets with value lasting many years. Human capability barely appears on the balance sheet. Supplier and customer financing can support the operating capital of the business. A local subsidiary may rely on intellectual property created elsewhere by a parent company or licensor.

A system-level view therefore asks a broader set of questions. Who financed every material asset required to generate the cash flows? Where was the scientific and technical capability created? Who bears the cost of experimentation and failure? Who owns the resulting intellectual property?

Those questions help separate businesses that are highly efficient users of capital from businesses that are also creating the scarce assets on which future rents will depend.

The benchmark shows similar systemic returns built on different assets

A comparison of the leading US Nasdaq companies with ten large Indian counterparts across ROIC, growth, cash flow, cost of capital and sustainability produces an interesting result. The analysis also introduces a directional measure of systemic ROIC, which capitalises innovation using a proxy equal to three times current R&D intensity and applies normalised tax rates of 21% in the US and 25% in India. It is intended as a comparative diagnostic rather than a precise measure of economic return. The main limitations are summarised in the methodological note below.[7]

The first result is striking. Median reported ROIC is about 21.1% for the Indian group, compared with 25.8% for the US frontier cohort. Once R&D is treated as an investment rather than a current expense, however, median systemic ROIC converges sharply: roughly 18.9% in India and 19.3% in the US.

The Indian companies are generating strong current returns from a relatively modest stock of internally created technological capital. Their economics are already highly efficient before substantial innovation capital is placed on the balance sheet. The US frontier companies, by contrast, carry a much larger burden of internally financed technological creation while still producing comparable underlying returns.

Source: [7]

The deeper difference lies in the asset base. Median R&D intensity is about 13% for the US cohort and 1.2% for the Indian group. Median three-year revenue growth is roughly 20.6% in the US cohort against 8.5% in India, while operating margins are about 35.6% versus 20.5%.

The US companies devote a much larger share of current income to creating proprietary intangible assets while also producing faster growth and higher margins. The Indian group generates strong returns through coordination, distribution, implementation, working-capital discipline and established physical systems.

Company cases reveal where the economic asset sits

TCS is perhaps the clearest Banianomic case in the information age. Its economic engine organises skilled human capital around enterprise software, cloud infrastructure, databases and customer systems whose underlying intellectual property is largely owned elsewhere. Formal R&D intensity is about 1.1%, yet the workbook still shows exceptionally high systemic ROIC.[8]

The productive asset sits in organisational capability: recruiting, training, process discipline, client relationships, systems integration and reliable execution at global scale.

HUL illustrates a different boundary. It combines global Unilever brands, formulations and product science with Indian manufacturing, localisation and distribution. The local entity can therefore produce very high capital efficiency while relying on intellectual assets financed partly outside its own accounting perimeter.

Maruti shows a similar structure through technology accumulated across the wider Suzuki system. M&M carries more of the engineering burden internally through product platforms, EV architecture, software and powertrains. A static return-on-capital comparison can therefore favour a company whose technology capital sits elsewhere in the corporate system.

Reliance, Bharti Airtel and L&T show another form of strength: system integration at scale. Reliance builds and commercialises enormous systems. Airtel integrates spectrum, equipment, towers, software and distribution. L&T coordinates technology suppliers, contractors, engineering and capital across highly complex projects.

Their advantage lies in orchestration. Much of the scarce technological primitive is purchased, licensed or integrated into a larger commercial system.

The US frontier cohort concentrates more of that scarce asset inside the firm. Nvidia owns semiconductor architecture and CUDA. Microsoft owns operating systems, enterprise software and cloud platforms. Alphabet owns search, data infrastructure and AI systems. Meta owns large-scale attention, recommendation and social-network infrastructure. Broadcom combines semiconductor intellectual property with mission-critical software.

These assets can often serve additional customers without a proportional increase in inventory, labour or physical distribution. That gives successful frontier firms a different relationship between capital creation, scale and margin.

Source: [7]

The scatter shows why ROIC alone is insufficient. TCS and Nvidia both sit near the high-return end of the sample, yet they arrive there through very different economic architectures. TCS monetises human capital, coordination and global delivery. Nvidia combines proprietary architecture with rapid growth and an expanding software ecosystem.

Amazon and Tesla provide an equally useful warning. A frontier position does not guarantee attractive current returns. Technological ambition can coexist with weak or volatile economics for long periods.

The capability gap is concentrated in IP ownership and J-curve tolerance

Execution capability is strong on both sides. The separation appears elsewhere: in ownership of scarce technological assets and in the willingness to finance long periods of uncertain payoff.

That is the more useful diagnosis. Indian corporate competence is not the problem. Many Indian companies are exceptionally good at execution, capital discipline, distribution and scaling. The weaker dimensions lie further upstream: creating proprietary technology, owning more of the intellectual property and tolerating the long J-curves required to build those assets.

Several Indian companies already show signs of transition. Sun Pharma has moved further into specialty and proprietary pharmaceutical assets. HAL carries a substantial aerospace and defence engineering burden. M&M is developing more of its own vehicle architecture, EV platforms and software. HCLTech has expanded into engineering and software products. Bajaj Auto increasingly combines manufacturing and distribution strength with proprietary product development and EV engineering. Reliance has also begun investing more heavily in technology platforms and new industrial capabilities.[9]

This transition can make the accounting look worse before the economics improve. Engineers are hired before products generate revenue. Development programmes fail. Payback periods lengthen. Capital becomes harder to recover. Reported returns can fall even while technological capability is increasing.

That creates a harder task for investors. They have to distinguish between lower returns caused by capital destruction and lower returns caused by the creation of an asset that may later support pricing power, reinvestment and strategic control.

The key questions therefore become forward-looking. What scarce asset is being created? Who will own it? How long can capital be reinvested into it? What will give the eventual product pricing power? How much failure can the balance sheet absorb before the capability becomes commercially valuable?

These questions matter more as a company moves from operating existing systems efficiently towards creating the technologies, intellectual property and platforms on which future economic rents will depend. In the companion workbook, Sun Pharma records the highest transition score, followed by M&M, HCLTech, Bajaj Auto, HAL and Reliance.

Source: [11a]

Using a broader top-ten corporate sample rather than the most R&D-intensive firms reduces selection bias without changing the hierarchy. Among companies with separately identifiable disclosure, the median R&D burden is roughly 10.7% in the United States, 4.0% in China and about 1.0% in India.[11a]

China converted manufacturing scale into technological ownership

China is the most useful bridge case because its industrial ascent began close to the low-cost manufacturing and assembly model. Its progression can be read as a learning curve. Labour and assembly created manufacturing scale. Scale produced dense supply chains and process knowledge. Engineering capability accumulated inside those systems. Policy and corporate investment then pushed selected sectors towards indigenous product architecture, intellectual property and standards.

During the 1980s and 1990s, China absorbed foreign direct investment, imported equipment, operated joint ventures and became a global contract-manufacturing base. Manufacturing itself became a learning system. Repetition produced quality systems, supplier density, engineering talent and exposure to international production methods.

By the mid-2000s, Chinese policy increasingly treated dependence on foreign technology and low-value production as strategic weaknesses. The 2006 science and technology plan placed greater emphasis on indigenous innovation, the absorption and reworking of imported technology, and higher national R&D intensity.[10]

Source: [11]

Table 4. China’s progression from assembly towards ownership of architecture and standards. Sources: China science-and-technology policy and company disclosures [10] [12].

Huawei, BYD and CATL show how this progression can happen inside companies. Huawei moved from reseller and integrator into switching technology, telecom equipment, standards, chips, operating systems, cloud and AI. BYD accumulated capability across battery chemistry, power electronics, motors, semiconductors, vehicle architecture and software. CATL moved from battery manufacturing scale into cell chemistry, battery-management systems, fast charging, sodium-ion technology, manufacturing processes and recycling. Manufacturing remained a source of advantage throughout. Intellectual property accumulated on top of it.[12]

China industrialised the J-curve

The climb was often financially inefficient. It produced bankrupt companies, excess capacity, duplicated factories, bad loans, subsidy losses and destructive price wars. Yet even failed capital sometimes left behind useful assets: engineers, suppliers, factories, patents, process knowledge and experienced managers.

China effectively treated part of capital destruction as tuition for industrial capability.

Table 5. Three ways of financing the J-curve. Author synthesis.

This model carries its own danger. Technological capability combined with extreme overcapacity can drive industry ROIC below WACC. Solar and several other sectors show what happens when capability accumulation outruns scarcity and capital discipline.

Technological ownership alone does not guarantee attractive economics. Durable frontier returns still require scarcity, pricing power and disciplined reinvestment.

Catch-up economics eventually reaches a productivity ceiling

Banianomics is especially powerful during catch-up development because the required machines, technologies, business models and organisational methods already exist somewhere else. The economic task is acquisition and execution: identify what works, buy or license it, assemble it, localise it, finance it, distribute it and scale it.

These capabilities can build ports, highways, telecom networks, cement plants, refineries, banks, consumer brands, generic pharmaceutical companies, software-services businesses and increasingly sophisticated manufacturing systems.

They can take an economy a very long way.

As an economy becomes richer, however, the easier productivity gains begin to narrow. Labour has already moved out of agriculture. Informal activity has become more organised. Machines have replaced many manual processes. Infrastructure gaps have closed. Domestic markets have matured. Wages and land costs rise. Labour-cost arbitrage weakens. More capital flows into activities that are already well supplied, pushing incremental returns lower.

The development problem gradually shifts towards originating productivity through better technology, deeper human capital and accumulated knowledge.

This is where a system centred on execution begins to face a ceiling. A conglomerate can move from cement into logistics, power, airports, data centres and renewable energy while retaining much of the same economic structure: large physical investment, formidable project execution and substantial reliance on technology created elsewhere.

These can still be excellent businesses. The strategic question is where the scarce bottleneck sits and who captures the highest rent.

A modern data centre makes the hierarchy visible. Local capital funds land, electricity, buildings, cooling systems, network connections, financing and customer acquisition. Further upstream, technological rents may accrue to Nvidia’s accelerators, TSMC’s fabrication, ASML’s lithography, Arm’s architecture, Synopsys and Cadence design tools, and the software and cloud layers controlled by Microsoft, Google or Amazon.

Physical investment and rent capture can sit in different parts of the same value chain.

Source: [14]

The middle-income trap can therefore be read partly as a productivity trap. Physical capital usually faces diminishing returns as obvious infrastructure gaps close. Knowledge has stronger replication economics. A semiconductor architecture, drug molecule, algorithm or software platform can be reused across millions of units with limited additional intellectual cost.[14]

Advanced economies therefore accumulate a growing share of capital in patents, software, algorithms, proprietary processes, scientific knowledge, brands, standards, datasets and organisational systems.

Technological ownership also shapes national income. Workers can become highly productive while operating machinery, software and platforms created abroad. Part of the resulting productivity surplus still flows to whoever controls the patents, components, standards and ecosystems.

The value hierarchy can be read as a progression from labour arbitrage through manufacturing and engineering towards intellectual-property and platform rents. Economies concentrated lower in that hierarchy can become large, sophisticated and highly capable while preserving a persistent income gap with economies that control more of the upper layers.

Source: [15]

Fragmenting globalisation raises the cost of external dependence

China used the roughly twenty-year window between WTO accession in December 2001 and the geopolitical turn around 2020 to absorb capital, technology and know-how, build manufacturing scale, deepen supplier ecosystems and push selected industries towards technological leadership.[16]

Indian companies operated through much of the same global window with labour-cost advantages, a rapidly expanding domestic market, access to foreign technology, attractive returns on capital and increasingly deep capital markets. Company-level incentives often rewarded localisation, distribution, infrastructure, financial engineering and regulatory access more quickly than a twenty-year programme of technological accumulation.

Those incentives made economic sense at the firm level. Their aggregate consequences become more important as globalisation fragments.

Today, industrial policy has returned. Technology is increasingly securitised. Supply chains are being reorganised. AI is moving towards more autonomous systems. Resilience now requires diversification across technologies, suppliers, markets, capital and intellectual property.[17]

It also demands a harder test of corporate strength. High returns need to be traced back to their source. Are they produced by scarce capabilities the firm controls? Do they depend on technology, capital or geopolitical conditions supplied by others? How durable are those advantages if the external environment changes?

Political language around resilience, self-sufficiency and Atmanirbharta should therefore be judged against the structure of capital allocation. Domestic manufacturing capacity matters, but strategic autonomy also depends on ownership of critical science, engineering, software, standards and intellectual property. A country can deploy modern technology at enormous scale while remaining dependent on others for the technologies that define the frontier.

That dependence also weakens bargaining power. Trade negotiations and international diplomacy are shaped partly by what a country controls and what it cannot easily replace. Technological dependence therefore becomes an economic constraint and, eventually, a strategic one.

A simple household experiment makes the point. Walk around an Indian home and look at the technology being used: phones, laptops, operating systems, processors, routers, cloud services, televisions, appliances and increasingly the software embedded inside them. Much of the underlying technology, intellectual property or critical componentry ultimately comes from the United States, China, East Asia or Europe. India may assemble, distribute, finance and consume these products at enormous scale while owning relatively little of the technological architecture underneath them.

This brings us back to Reliance, Adani, Tata and the other large Indian corporate houses. The issue is not that they are young businesses that have simply had insufficient time to climb the learning curve. Many have existed for generations. They have become extraordinarily capable at the activities their economic environment rewarded: execution, distribution, project management, capital allocation, regulatory navigation and the scaling of proven models. They have been much slower to move upstream into the creation of the scarce technological assets themselves.

Banianomics helps explain why. Its natural instinct is continuous improvement within a proven economic model: become more efficient, turn capital faster, expand distribution, reduce costs and compound what already works. That can produce exceptional businesses for decades. It is less naturally suited to creative destruction. Frontier capitalism sometimes requires a company to undermine its own existing economics, tolerate years of lower returns and spend heavily on technologies that may never work. The organisation has to risk a profitable present to create an uncertain future. Under Banianomics, the instinct is often to keep getting better at what already works until the economic cycle, a new technology or an external competitor changes the game. That is precisely where the strength of the model can become its constraint.

Investors should value reinvestment runway alongside current ROIC

Growth = Reinvestment Rate × Incremental ROIC

Current ROIC describes the economics of the existing capital base. Long-term value depends on what happens to the next unit of capital: how much can be reinvested, what return that new capital can earn, and how long those returns can remain above the cost of capital.[18]

A mature distribution network may report an exceptional ROIC while offering limited scope to deploy additional capital at the same rate. A company building proprietary technology may report a lower current ROIC while creating an asset capable of absorbing capital at attractive returns for many years.

The relevant variables are therefore incremental ROIC, reinvestment capacity and duration.

Consider two illustrative companies, each earning ₹100 of current NOPAT and seeking to increase NOPAT by 20%, or ₹20.

*Assumes, for illustration, that current NOPAT broadly approximates operating cash available for reinvestment before other uses.

Table 6. Same earnings growth, very different capital burden.

Required reinvestment = Increase in NOPAT ÷ Incremental ROIC

The integrator requires ₹80 of additional capital to generate ₹20 of new NOPAT. The high-incremental-ROIC IP business requires only ₹25.

Both produce the same increase in earnings. Their capacity to compound is very different.

The integrator must commit most of its internally generated resources to support the next round of growth. The IP business retains substantially more cash after funding the same increase in NOPAT. That surplus can finance additional products, acquisitions, new markets, distributions or further experimentation.

This advantage becomes especially powerful when the underlying asset can be reused without a proportional increase in physical capital. Software, intellectual property, technical architectures, platforms and other scalable intangible assets can sometimes support large increases in revenue without requiring equivalent increases in inventory, factories, distribution infrastructure or labour.

High incremental ROIC alone is insufficient. The opportunity must also be large enough to absorb meaningful amounts of capital. A company capable of earning 80% on only a small amount of new investment may create less value than one capable of reinvesting billions for decades at 25% or 30%.

The investor therefore needs to ask three questions: How much capital can be reinvested? At what incremental return? For how long?

Two valuation errors follow

The first is extrapolating mature ROCE too far into the future.

Supplier finance, underpenetrated distribution, low initial capital requirements and exceptional asset utilisation can produce current returns of 35% or 40%. Those economics may weaken as markets mature, distribution saturates and each additional unit of capital becomes less productive. Historical ROCE can therefore overstate the future opportunity set.

The second error is treating capability investment as ordinary deterioration.

Heavy R&D, proprietary software, engineering platforms and vertical integration can depress reported ROCE long before the resulting asset generates meaningful revenue. Engineers are hired before the product exists. Development programmes fail. Payback periods lengthen. Capital becomes harder to recover.

The relevant question is whether this spending is accumulating a scarce capability with future pricing power, strategic control and a long reinvestment runway.

Repeated technical failure with little accumulated capability deserves one judgment. An investment programme that steadily increases ownership of the technology stack deserves another.

This distinction matters particularly for Indian companies moving from trader → integrator → designer → technology owner. Their accounting efficiency may weaken during the transition even as the underlying economic capability improves.

The evidence lies in the direction of travel: R&D intensity, internal ownership of technology, patent and engineering quality, gross-margin development, control over product architecture, dependence on external suppliers, reinvestment runway and the ability to convert technical capability into economic scarcity.

For investors, the central question is therefore broader than current ROIC: What is the company becoming capable of earning on the capital it has yet to deploy?

India needs a parallel frontier-capital operating system

Banianomics helps explain how Indian capitalism became highly effective at preserving scarce capital, exploiting visible opportunities, coordinating complex systems and compounding through commercial execution. It can produce outstanding shareholder returns and carry a poor country a long way through diffusion, localisation and scale.

Those strengths remain valuable.

The next development challenge is knowledge accumulation.

Frontier technology requires capital that can remain illiquid and loss-making for years. It must absorb technical failure, finance uncertain science and tolerate the destruction of existing profit pools when new technologies emerge. That requires more than venture capital. It requires universities, public research, deep capital markets, corporate R&D, defence and industrial programmes, patient owners and portfolios capable of carrying long J-curves.

India therefore needs an additional operating system for capital.

Banianomics can continue to supply the disciplines of commercialisation, deployment, working-capital control and execution. Frontier capital must finance scientific risk, proprietary technology, intellectual property and new market creation.

The strongest future companies will combine both traditions. They will retain merchant discipline in the use of capital while developing the patience to create scarce technological assets.

The same principle applies at the national level. India does not need to become less commercially disciplined. It needs to become more willing to finance forms of capital whose value cannot be seen clearly at the moment the money is committed.

Perhaps the hardest discipline for any successful system is recognising when yesterday’s strengths have begun to create new vulnerabilities.

Resilience starts with diversification at every level, and with avoiding the oldest mistake in strategy: mistaking your own odour for perfume.

Methodological note

The company ratios are analytical snapshots rather than audited reconciliations. Systemic ROIC is a directional diagnostic built with normalised tax assumptions and a proxy for hidden R&D capital. ROCE definitions vary across companies and providers; external IP and supplier/customer financing are difficult to measure precisely; cross-country R&D intensity is affected by sector mix; and the R&D–income charts show association rather than causality.

Notes and sources

1. [1]Claude Markovits, The Global World of Indian Merchants, 1750–1947: Traders of Sind from Bukhara to Panama (Cambridge University Press, 2000), especially chapters 1 and 5 on South Asian merchant networks and business organisation. DOI
2. [2]Thomas Sowell, Migrations and Cultures: A World View (Basic Books, 1996), chapter “Overseas Indians”.
3. [3]Mohnish Pabrai, The Dhandho Investor: The Low-Risk Value Method to High Returns (Wiley, 2007), chapter 1, “Patel Motel Dhandho”.
4. [4]Fernand Braudel, Civilization and Capitalism, 15th–18th Century, Vol. III: The Perspective of the World (University of California Press, 1992), especially the discussion of world-economies and their hierarchy of zones.
5. [5]Joseph A. Schumpeter, Capitalism, Socialism and Democracy (Harper, 1942), ch. VII, “The Process of Creative Destruction”.
6. [6]Aswath Damodaran, NYU Stern, “High Growth Period” and “Terminal Value and Excess Returns,” on reinvestment, excess returns and competitive-advantage duration. High-growth period | Excess returns
7. [7]Companion Banianomics benchmark workbook, 10 September 2026. Eight US frontier companies and ten Indian comparables; inputs include company filings and standardised operating metrics. Systemic ROIC is a comparative diagnostic rather than an audited company measure.
8. [8]Company disclosures and the companion benchmark workbook. See TCS FY2026, HUL FY2026 and M&M FY2026.
9. [9]Company disclosures and the companion benchmark workbook, including Sun Pharma FY26 and M&M FY2026. Transition scores are analytical judgments rather than audited measures.
10. [10]State Council of the People’s Republic of China, Outline of the National Medium- and Long-term Program for Science and Technology Development (2006–2020); see also OECD Science, Technology and Innovation Outlook 2023 on China’s indigenous-innovation drive.
11. [11]Sources: World Bank World Development Indicators / UNESCO Institute for Statistics, “Research and development expenditure (% of GDP)”; India Department of Science & Technology, Research & Development Statistics 2025–26; China National Bureau of Statistics, 2024 R&D expenditure. World Bank | India DST | China NBS
12. [11a]Analytical comparison assembled for this essay from the latest full-year public disclosures of the 10 largest listed non-financial companies by market capitalisation in each country. Headline figure uses the median reported R&D intensity among companies with separately identifiable current-year R&D disclosure: United States 10.72% (9/10 disclosed), China 3.99% (10/10), India 1.03% (5/10). Principal filings include Apple FY2025, Alphabet FY2025, Microsoft FY2026, Broadcom FY2025, Meta FY2025, Tesla FY2025, Eli Lilly FY2025, Micron FY2025, and company disclosures for Tencent, Alibaba, CATL, China Mobile, PetroChina, Reliance, TCS, Infosys, Sun Pharma and Titan. Sector composition is intentionally not normalised because it forms part of the argument about each market’s corporate structure.
13. [12]Company disclosures: Huawei 2025; CATL 2025; BYD Annual Report 2024 / sustainability disclosures on R&D and vertically integrated technology development.
14. [13]World Bank, World Development Report 2024: The Middle-Income Trap. The report distinguishes an investment-and-infusion catch-up phase from the later requirement to add innovation and push the technological frontier. Source
15. [14]World Bank World Development Indicators. R&D intensity uses indicator GB.XPD.RSDV.GD.ZS (UNESCO UIS); GDP per capita values are PPP-based. Latest available country years differ, so the figure is a descriptive cross-section rather than a causal estimate. Source
16. [15]World Bank World Development Indicators / UNESCO UIS. GDP per capita is shown in PPP constant 2021 international dollars and R&D intensity as gross domestic R&D expenditure as a share of GDP. The plotted trajectories are approximate visual summaries and do not establish causality. Source
17. [16]World Trade Organization, China accession record. China became the WTO’s 143rd member on 11 December 2001. Source
18. [17]OECD, Science, Technology and Innovation Outlook 2023, on the return of strategic industrial and technology policies, China’s indigenous-innovation programme and intensifying competition around critical technologies. Source
19. [18]For the valuation logic, see Aswath Damodaran, NYU Stern materials on high-growth periods, reinvestment and excess returns. The numerical example is illustrative rather than a forecast. Source