Singapore's Grant for First-Time Founders Got 65 Applications Last Year. It Approved 55

SINGAPORE, 8 AUG 2026 — Singapore's grant for first-time founders received 65 applications in the whole of 2025. It approved 55 of them.

An 85 per cent approval rate on a two-digit application count is not a selective programme. It is a programme almost nobody applies to.

The numbers

Mr Patrick Tay Teck Guan asked how many applications the Startup SG Founder programme received in 2025, how many succeeded, and the breakdown of successful applicants by age, gender and grant amount.

The written reply from Minister for Trade and Industry (Energy and Industry) Dr Tan See Leng was unusually complete.

Fifty-five grants at roughly S$50,000 each puts the programme's annual outlay somewhere near S$2.75 million. For a national scheme in a country that describes entrepreneurship as an economic priority, that is a small number attached to a small number.

The eligibility rules explain the volume

A second reply the same week, to Mr Ng Shi Xuan, sets out who may apply, and it narrows the field considerably.

Each application must come from two applicants. The first must not have previously registered a private limited entity with ACRA. The private limited entity used for the application must have been registered in Singapore for less than six months at the time of applying.

Read together, those conditions exclude anyone who has ever incorporated anything, and anyone whose company is more than six months old. A founder who registered a company two years ago and let it lapse is out. A founder who incorporated eight months ago and spent that time building is out.

The ministry has loosened this once. Previously both applicants had to be first-time founders; EnterpriseSG refined the criterion in 2024 so only the first must be, which lets a first-timer pair with an experienced founder and draw on their expertise and networks. The reply says this has helped create a more robust startup ecosystem and that the arrangement continues.

What the design is actually for

The programme's stated objective explains its constraints.

The scheme supports first-time founders who may lack the experience, networks and resources to pursue entrepreneurship. This is not a startup subsidy. It is an access programme for people who would not otherwise start, with eligibility rules designed to find them.

Judged that way, an 85 per cent approval rate is a feature. If the gate is at eligibility, then most people who clear it should succeed, and a low rejection rate means the criteria are doing the filtering rather than a committee.

The awkward part is the denominator. Sixty-five applications in a year of roughly 250 working days is about one application every four days, nationally. Whatever the design intends, it is reaching very few people.

Who is getting it

The demographic breakdown requires careful reading.

Two thirds of successful founders were under 40, and the range ran from 23 to 69, so the scheme is not exclusively young. Sixty-nine per cent were male.

That last figure invites a comparison the reply does not enable. Without the gender split of the 65 applications, it is impossible to know if that 69 per cent figure reflects the applicant pool or the selection process.

On a base of 55 grants, the difference between 69 per cent and parity is about ten people, which is worth remembering before treating the percentage as a structural claim.

What the number suggests

Sixty-five applications can mean several things and the reply does not distinguish between them.

The low count could mean the rules are working, filtering for a cohort that is simply small. Or the six-month incorporation window could be too tight for founders who only seek grants after setting up a bank account. Or awareness is low. Or S$50,000 just is not enough to convince someone to start a company.

Each explanation implies a different fix. Telling them apart would require data from people who considered applying and did not — data that grant programmes never collect.

What S$50,000 buys

The grant size is worth holding against the thing it is meant to enable.

Roughly S$50,000 is a few months of one founder's forgone salary, or a modest prototype build, or a year of very lean operating costs for two people who are not paying themselves. It is not a seed round and does not pretend to be.

For the intended cohort, people who lack the networks and resources to begin, S$50,000 can be decisive. Their constraint is the first step, not scale. For anyone already able to raise privately it is close to irrelevant, which is consistent with the eligibility rules pushing that group out.

The design is coherent. The application count, however, raises the question of whether it is reaching the people it describes.

The measurement problem underneath

Access programmes are unusually hard to evaluate, and this one illustrates why.

The published outputs are applications, approvals and demographics of the approved. The real measures of success are counterfactuals: how many of the 55 would have started anyway, and how many qualified founders never applied. Neither is observable in the data released, and the second is not observable at all without deliberately going to look for it.

An approval rate cannot substitute. A programme that approved everyone and a programme that received applications only from people certain to succeed would look identical on these numbers.

Two applicants, one first-timer

The pairing requirement is the most distinctive feature of the design and the least discussed.

Every application needs two people, and since 2024 only the first must be a first-time founder. The rationale is to let a first-timer work with an experienced founder, a sensible response if inexperience, not capital, is the main obstacle.

It also imposes a real precondition. An aspiring founder who has not yet found a co-founder cannot apply at all, and finding one is itself a networks problem — the same problem the scheme exists to solve. The rule assumes the applicant has already cleared the hurdle the programme is designed to help with.

Whether that filters out serious candidates or usefully tests commitment is not something the published data can settle, since it counts only those who got as far as applying.

What a comparison would need

The most useful thing about these figures is also what makes them hard to interpret: they arrive without context.

A reader wanting to know whether 65 is low needs the same figure for previous years, the number of new private limited companies registered in Singapore over the same period, and ideally the count for comparable schemes elsewhere. None of the three is in the reply, and only the second is straightforwardly public.

That is not a criticism of the answer, which gave precisely what was asked. It is an observation about how programme data reaches the public in this form: one year at a time, in response to whoever thought to ask, with no series to place it in.

What to watch

Two things.

The first is whether the six-month registration window is revisited. It is the most mechanical of the constraints and the one most likely to exclude founders the scheme is designed for.

The second is whether application counts will be published alongside approvals. The rule change in 2024 was meant to loosen requirements; without application data, we cannot know if it worked.