The wealth gap is widening. The answer isn’t more handouts – it’s a stake

Here’s a statistic on wealth inequality in Singapore that might make you sit up: The average wealth held by the top 20 per cent of households is about $5.3 million per household, more than 18 times that of the bottom 20 per cent, which is $293,000.

The average wealth of the top 20 per cent also dwarfs the average wealth held by all the percentiles of the bottom 80 per cent added up, which totals around $3.5 million. This does not mean the top fifth owns more than the other four-fifths combined, since the latter group is far larger. But it signals a meaningful and troubling disparity in wealth per household.

The source is an Occasional Paper on Income Growth, Inequality, and Social Mobility Trends issued by the Ministry of Finance (MOF) in early February. The MOF candidly acknowledges that the wealth of the richest group may be underestimated because people tend not to disclose sensitive financial information or have difficulty recalling details – which means inequality may actually be higher than the estimates suggest.

Singapore’s overall Gini coefficient for wealth stands at 0.55, meaningfully higher than the income Gini coefficient of 0.38 after taxes and transfers. (The Gini coefficient runs from 0 to 1; the higher the figure, the greater the inequality.) The shift towards higher wealth taxes seen in recent budgets is, in that light, entirely logical.

Capital Is pulling away from labour

But there is a deeper structural force at work, one that wealth taxes alone cannot address. The mainstreaming of artificial intelligence since 2022 has accelerated a trend that was already under way: the returns on capital – financial assets in particular – are running far ahead of wage growth or even economic expansion.

Since those who are already wealthy hold the lion’s share of financial assets, they are compounding their advantage compared with the rest of the population, which depends primarily on earned income.

The economist Thomas Piketty, in his landmark book Capital In The Twenty-First Century, captured this dynamic in a simple but powerful formulation: r > g. When the rate of return on capital (r) exceeds the rate of economic growth (g), inequality will inexorably rise. The data from the past three years show that this is exactly what has been happening.

In the US, where AI first took off, the trend is dramatic.

From 2023 to 2025, the S&P 500 – a reasonable proxy for returns on capital – rose by 86 per cent. Over the same period, nominal wage growth came to around 12.6 per cent (less in real terms, after accounting for inflation), while real gross domestic product (GDP) grew a cumulative 8 per cent.

In short, the S&P 500 outpaced wages by almost seven times and GDP by more than 10 times.

In Singapore, a similar story played out, if less dramatically. Including dividends reinvested, the Straits Times Index (STI) rose a cumulative 62.8 per cent from 2023 to 2025, while nominal wages grew 16.5 per cent and real GDP 10.5 per cent.

Capital returns from the STI outpaced nominal wages by roughly 3.8 times and real GDP by almost six times.

What is happening, in essence, is a massive redistribution – not from rich to poor, but in the opposite direction: from labour to capital, from wages to profits, from workers to shareholders.

And this trend may accelerate in the AI era, which, despite its promising long-term benefits, is likely to displace many workers in the short term, compress wages for white-collar workers transitioning to new roles, and concentrate the astronomical profits of technology companies in the hands of those who own them.

A floor is not enough

One policy response commonly advocated to deal with technological disruption is Universal Basic Income or UBI – a regular cash payment to every citizen regardless of employment status. Silicon Valley executives have been among its most vocal champions.

But UBI, whatever its merits as a safety net, does not address the problem of rising inequality. As Nobel laureate Michael Spence put it: “It just puts a floor under incomes – so you get most people sitting on the floor while a select few capture the astronomical wealth generated by capital.”

A more promising response is Universal Basic Capital or UBC – giving everyone, not just the wealthy, a stake in the assets that are generating outsized returns.

As James Manyika, a senior vice-president at Google, has argued: “It’s crucial that we have more people participating in the capital income pathway because, while labour income remains the most important for the majority of people, capital income is a bigger and bigger part of where the value is going.”

The distinction matters. UBI redistributes income after the fact. UBC builds wealth from the ground up – harnessing the power of compounding so that those who currently have no financial assets begin to accumulate them early, and over time. Critics will note that the benefits of UBC accrue over decades rather than immediately, making it a poor remedy for current poverty.

That is a fair point – but it argues for UBC complementing, rather than replacing, existing social support schemes, not for abandoning the idea.

Putting UBC into practice

The only fully operational example of UBC today is Alaska’s Permanent Fund Dividend, which has paid annual cash distributions to every resident from oil revenues since 1982. But several other models are emerging.

The US has proposed “Trump Accounts,” under which every child under eight would receive a US$1,000 (S$1,300) seed investment, with families able to contribute up to US$5,000 a year into an index fund. Withdrawals would be permitted from age 18 onwards, for education, a business or a home.

Germany has come up with “early start pension accounts”, crediting children aged six to 17 with €10 (S$14.70) a month, which parents and later the beneficiaries themselves can top up and invest in diversified products such as exchange-traded funds; the funds can be withdrawn only at retirement, when decades of compounding would have made them substantial.

OpenAI has proposed a Public Wealth Fund that would give every citizen a stake in AI-driven growth, potentially seeded with equity contributions from AI companies.

Singapore’s opportunity

Singapore is unusually well-positioned to act on this idea because the infrastructure for UBC already exists in embryonic form. The Central Provident Fund (CPF) is a sophisticated, trusted savings and investment architecture. With targeted adaptation, it could serve as the vehicle for a genuine UBC scheme.

The existing CPF structure could remain intact. But new accounts within the system could be created and opened to all citizens from an early age – not only those in employment, as is currently the case.

For citizens without savings, the Government could seed these accounts via interest-free loans, repayable later from the account’s own returns or from future earnings.

The forthcoming life-cycle investment scheme to be introduced in 2028, which adjusts asset allocation as the account holder ages, provides a further ready-made mechanism that could be extended to UBC accounts.

The goal is simple to state, even if the implementation requires care: Every Singaporean, from as early as possible in life, should have some exposure to financial assets that compound over time.

This would not be a handout nor a floor, but a stake in the same wealth-generating mechanism that has been working so powerfully, and so unevenly, for those already at the top.

The wealth disparity will not close by itself. In the age of AI, it is more likely to widen. A UBC can help narrow the gap.