The Missing Middle of Indian Enterprise

Pritiraj BrahmaFounding Partner, Pritiraj & Partners · 9 April 2026

Only 14% of India's MSMEs have formal credit, against a gap estimated between ₹25 and ₹30 lakh crore. Indian firms stay small because staying small is the rational response to the system around them.

ProvenancePublished 9 April 2026; revised 29 August 2026 to add two figures, a fifth and sixth finding on the delayed-payment stock and the registered size distribution, and to remove a claim the essay’s own counter-case had weakened. Credit gap estimates are Deloitte’s State of Financial Services in India (position as at March 2025) and a SIDBI study; they are modelled quantities and are reported as a range. Registration figures are the Ministry of MSME’s Udyam data on data.gov.in as at June 2026, alongside the Ministry’s Udyam Registration Portal and Udyam Assist Platform total to 28 February 2026 — the two counts differ and both are stated. Delayed-payment figures are from the Ministry of MSME Annual Report 2025-26 (position as at 31 December 2025) and the Economic Survey 2025-26. Bank credit access shares are from NITI Aayog with the Institute for Competitiveness, 2020 to 2024. Outstanding MSME credit is as stated by the Minister for MSME in the Rajya Sabha, August 2026. The MSME Development (Amendment) Bill, 2026 is described as passed by both Houses in August 2026; commencement depends on notification and had not been confirmed at the time of revision.

India does not have a shortage of entrepreneurs. It has a shortage of firms that grow. The reason is visible in the credit data, and it is not a shortage of capital in the economy.

Deloitte’s State of Financial Services in India report finds that only 14 per cent of India’s micro, small and medium enterprises have access to formal credit, leaving the majority dependent on informal and usurious financing. It estimates the MSME credit gap at around ₹25 lakh crore as of March 2025, and notes that measured against the sector’s contribution to GDP and a healthy credit-to-GDP ratio, the gap could be well over ₹50 lakh crore. A SIDBI study puts the addressable gap at about 24 per cent of demand, or roughly ₹30 lakh crore, with formal debt supply from banks and NBFCs at about ₹34 lakh crore against addressable demand of about ₹64 lakh crore.

Estimates differ because definitions of “addressable demand” differ, and this Review does not propose to adjudicate between them. The range itself — ₹25 to ₹30 lakh crore on conservative measures — is the finding. Set against it is a number the government states plainly: outstanding credit to MSMEs has risen from ₹10 lakh crore in 2014-15 to over ₹38.35 lakh crore, as the Minister for MSME told the Rajya Sabha in August 2026. Supply has more than tripled. The gap has not closed, because it was never principally a supply problem.

Six findings anchor this analysis:

1. The middle is not thin. It is a rounding error. On the Ministry of MSME’s Udyam data published on data.gov.in, 4,71,55,579 enterprises were registered as at June 2026. Of those, 98.9 per cent are micro, about 4.9 lakh are small, and 37,042 — in a country of 1.4 billion people — qualify as medium. For every medium enterprise India has registered, it has registered roughly 1,260 micro ones.
2. Access is improving from a very low base. NITI Aayog’s report with the Institute for Competitiveness records the share of micro and small enterprises accessing credit through scheduled banks rising from 14 to 20 per cent between 2020 and 2024, and medium enterprises from 4 to 9 per cent. Real movement; still a minority of firms.
3. Payments infrastructure reached firms that credit did not. Around 89 per cent of Indian adults now have financial accounts, and over 90 per cent of surveyed MSMEs accept digital payments — yet only about 18 per cent have taken a digital loan. India solved the transaction problem and left the underwriting problem largely intact.
4. Informality is both cause and effect. Roughly 35 per cent of surveyed MSMEs remain unregistered. An unregistered firm cannot access formal credit or schemes; a firm without credit has little incentive to bear the cost of registration. The loop closes on itself.
5. The largest single source of working capital stress is a receivable, not a loan. The Economic Survey 2025-26, cited in the Ministry of MSME’s own annual report, estimates that around ₹8.1 lakh crore is locked in delayed payments to MSMEs; a second published estimate puts it at ₹7.34 lakh crore. Against that, micro and small enterprises had filed 2,56,892 applications worth ₹55,244.31 crore on MSME Samadhaan to 31 December 2025 — and Facilitation Councils had disposed of ₹14,638.38 crore. The claimed sum is a fraction of the estimated stock, and the adjudicated sum a fraction of that.
6. The gap is worse for women-led enterprise. SIDBI estimates the credit gap at about 35 per cent for women-owned MSMEs against 24 per cent overall — a differential that persists even among women-led firms that do have some access.

Why firms choose to stay small

The standard explanation is that small firms lack collateral and records. True, and incomplete. The deeper issue is that Indian regulation is banded by size, and each band carries a step change in obligation.

Crossing a size threshold brings labour law obligations, statutory compliances, tax regimes and inspection exposure that arrive as a cliff rather than a slope. A firm just below a threshold faces a choice between staying below it or absorbing a discontinuous jump in cost and risk. Many rationally split into several smaller entities instead of growing into one larger one — which is why India has an enormous number of tiny firms, a reasonable number of large ones, and a conspicuously thin middle.

The revised classification notified with effect from 1 April 2025 raised the ceilings — micro to ₹2.5 crore of investment and ₹10 crore of turnover, small to ₹25 crore and ₹100 crore, medium to ₹125 crore and ₹500 crore — and Udyam now re-verifies a firm’s class automatically each year from its income tax and GST filings. Raising a threshold moves the cliff. It does not turn it into a slope. And automatic reclassification makes the edge sharper, not softer: a small enterprise that crosses ₹100 crore of turnover becomes medium, and in becoming medium it loses the statutory 45-day payment protection that covers only micro and small suppliers. Growth is penalised by the removal of a protection.

This matters because productivity, formal jobs, exports and the ability to service credit all live in the middle. MSMEs account for roughly 31 per cent of India’s GDP, 36 per cent of its manufacturing output and 41 per cent of its exports. A country of micro-enterprises generates livelihoods but not the productivity growth a demographic window requires.

What cash-flow lending changes, and what it does not

The most promising development is the shift from collateral-based to cash-flow-based underwriting, using the Account Aggregator framework, GST returns and digital payment histories. This is genuinely transformative for the firm that transacts digitally and files returns: its cash flow is now legible without a balance sheet or a mortgage.

Two cautions belong alongside the optimism. It reaches the registered and digitally transacting firm — which is, by construction, not the 35 per cent that are unregistered or the smallest micro-enterprises operating in cash. And underwriting on recent transaction history is inherently pro-cyclical: it lends generously into a good year and withdraws sharply in a bad one, which is the opposite of what a small firm needs from a lender.

The counter-case, honestly stated

Three objections deserve a hearing.

First, the credit gap is a modelled quantity, not a measured one, and the agreement between estimates is not corroboration. Every figure quoted — ₹25 lakh crore, ₹30 lakh crore, the ₹50 lakh crore upper case, the ₹64 lakh crore of addressable demand — is derived by asking what firms ought to be borrowing given the sector’s share of output or a benchmark credit-to-GDP ratio. That is a normative construction. It counts as unmet demand every firm that would not survive leverage, and lending to those firms destroys capital rather than releasing it. Some meaningful portion of the gap is rational credit rationing working as intended, and no estimate built this way can distinguish that portion from genuine exclusion. The delayed-payment stock in Figure 2 is a partial answer to this objection but not a complete one: a confirmed unpaid invoice is a measured receivable, yet ₹8.1 lakh crore of it is itself an estimate, and only ₹55,244 crore has ever been formally asserted by a supplier.

Second, the regulatory-cliff explanation is plausible and is offered without the evidence that would establish it. If firms are choosing to stay below size thresholds, they will bunch immediately beneath them, and bunching is directly testable against GST turnover distributions and factory registration data. This essay runs no such test, and the Udyam distribution in Figure 1 does not substitute for one: it records the classes firms fall into, not the choices they made near a boundary, and a base that is 98.9 per cent micro is equally consistent with a country whose firms are simply small. Absent the test, the cliff story competes with plainer candidates: a shortage of managerial depth, unreliable power and land, and contract enforcement so slow that firms vertically integrate rather than contract — a mechanism this Review sets out at length in its own work on judicial delay, where the inability to enforce a supply contract is named as a reason Indian firms stay small. The thin middle is probably overdetermined. Attributing it chiefly to threshold effects is an inference wearing the clothes of a diagnosis.

Third, phasing obligations in over three to five years does not cost nothing; it moves the cost. A glide path means the state must know, for every firm, which year of transition it occupies across labour, tax and inspection regimes simultaneously — and must resist the restructuring that any threshold-with-a-timer invites. That tracking burden lands on the same district and state machinery this Review describes as carrying a dozen mandates designed for one. The MSME Online Dispute Resolution portal is the cautionary case: given a ₹189 crore outlay and every new delayed-payment filing in the country, it disposed of seventeen cases worth ₹60.60 lakh in eight months. The reform is right in principle. It is free only if enforcement capacity is free, and the Review’s own position is that it is the scarcest input in Indian government.

One recommendation is immune to all of this, and it is the strongest in the list. A confirmed, dated public purchase order is a measured receivable rather than a modelled demand, and the state’s own payment delay is a documented cause of working-capital stress. That can be fixed without settling any of the arguments above.

What we would do

1. Replace regulatory cliffs with slopes. Phase obligations in over three to five years after a firm crosses a threshold, rather than applying them in full on day one — and start with the perverse one, under which a small enterprise growing past ₹100 crore of turnover forfeits the 45-day payment protection. Raising the ceilings, as the April 2025 classification did, moves the cliff; it does not remove it.
2. Make registration worth its cost. Registration should carry an immediate, tangible benefit — a credit line, a procurement preference, a faster refund — rather than functioning primarily as an entry into the compliance system.
3. Use public procurement as the credit event it already is. Government is the largest buyer from Indian MSMEs. A confirmed, dated public purchase order is excellent collateral, and delayed public payment is one of the largest sources of MSME working-capital stress. The MSME Development (Amendment) Bill, passed by the Rajya Sabha on 3 August and the Lok Sabha on 7 August 2026, requires central public sector enterprises to route MSME invoice settlement through the Trade Receivables Discounting System and makes awards recoverable as arrears of land revenue. Those are the right two instruments. Neither operates until commencement is notified, and the department that owns the notification should publish the date.
4. Publish the delayed-payment ledger, buyer by buyer. The Ministry already holds the Samadhaan case data: who was claimed against, for how much, and how long the Council took. Publishing it by buyer — central department, state department, public enterprise, private firm — would do more than any interest provision, because the sanction that works on a persistent late payer is the list.
5. Set an explicit women-led enterprise target with published data. A 35 per cent gap against a 24 per cent baseline is specific enough to be a departmental objective. It currently belongs to nobody.
6. Report credit-gap closure at district level. National aggregates hide the fact that credit access varies enormously by district. Udyam registrations are already published by district for all 783 of them; the credit series should be too, so states can direct guarantee and interest-subvention capacity where it actually binds.

India’s small firms are not undercapitalised because Indian banks lack money. Outstanding MSME credit has more than tripled in a decade, and the middle of the distribution is still 37,042 firms. They are undercapitalised because the system cannot see them, because the ones it can see have good reason not to grow into full visibility, and because the largest sum owed to them is not a loan anyone declined but an invoice somebody has not paid.