What Investors Actually Pay For

SGX RegCo wants every listed company to publish a dividend policy from January 2027, partly to narrow persistent valuation discounts. The evidence points elsewhere. Companies earning above their cost of equity trade at 1.63 times their net assets; the rest trade at a discount. Paying a dividend does not change that.

From 1 January 2027, every company listed on SGX must have a dividend policy and describe it in its annual report. SGX RegCo, the exchange’s regulator, says “a sizable proportion of issuers have chosen not to disclose their dividend policy.” That, it says, “can undermine investor confidence and may contribute to persistent valuation discounts.”

Tan Boon Gin, RegCo’s chief executive, set out the aim when the proposal was launched in April 2026. He told The Edge Singapore that RegCo was “calling on [companies] for more transparency to their investors.” A company that is not paying dividends because it needs to reinvest its profits “should say so. Investors can then express their view on whether they agree this is the best use of cash on the balance sheet.” The rule does not force any company to pay. It also “does not prescribe any specific dividend approach, payout ratio or target.”

Two other rules start on the same day. Companies must name the key performance indicators (KPIs) used to set executive pay, and explain how they link to long-term value. They must also keep an investor relations policy and website.

The dividend rule rests on a claim that can be checked: will a dividend policy lift valuations?

Only returns above the cost of equity earn a premium

Singapore investors are focused on returns. They pay more than net asset value for one kind of company only: one that earns more than its cost of equity (See Chart 1).

Return on equity (ROE) is a company’s net profit for the year divided by its shareholders’ equity, the money shareholders have in the business. The cost of equity is the return shareholders could expect from other investments carrying similar risk. If a company earns less than that, its shareholders would do better with their money elsewhere. For most SGX companies, the cost of equity is around 7% to 8%. This article uses 8%. Using 7% or 10% instead gives much the same result: companies earning above that level trade at 1.61 to 1.69 times their net assets, and profitable companies earning below it at 0.57 to 0.62 times.

On 1 October 2026, 615 counters were listed on SGX. Excluding REITs and business trusts, duplicate counters, suspended companies and those without the data needed leaves 479 companies, in three groups of almost equal size:

• Loss-making companies: 153 (31.9%);

• Profitable companies earning less than 8% on equity: 167 (34.9%);

• Profitable companies earning 8% or more: 159 (33.2%).

Companies earning 8% or more traded at a median 1.63 times net assets. Profitable companies earning less traded at 0.59 times. Loss-makers traded at 0.69 times, slightly higher, because heavy losses shrink net assets and push the ratio up.

Share prices tell the same story (See Chart 1). From January 2025 to October 2026, the median share price of companies earning 8% or more rose 25.8%. Profitable companies earning less rose 8.6%. The median loss-maker did not move. The pattern held in rising and falling markets alike.

In other words, the SGX companies that trade at a discount are those that earn less than their cost of equity.

Dividends do not buy a higher valuation

Paying a dividend does not lift a company out of that discount. A company that pays no dividend but earns 8% or more is valued at 1.22 times its net assets. A company that pays a dividend but earns less than 8% is valued at 0.60 times (See Chart 2). Investors pay twice as much for returns without a dividend as for a dividend without returns.

Within each group, dividend payers do trade a little higher. That is partly because payers tend to be larger companies, and larger companies tend to carry higher valuations. So the two effects need to be separated.

To do that, we used a multivariate regression: a statistical test that measures how much of the difference in valuation between companies each factor explains, while holding the others constant. We tested four factors: return on equity, whether a company pays a dividend, company size and sector. On 1 October 2026, together they explained 35.8% of the differences in valuation (See Chart 3):

• Return on equity: 19.2 percentage points;

• Company size and sector: 16.1 percentage points;

• Paying a dividend: 0.5 percentage points.

To check this was not a one-off, we repeated the test on SGX data for five other dates between January 2025 and June 2026. Return on equity explained between 13 and 25 percentage points each time. Dividends never explained more than 1.1.

The same test works on share prices. In Chart 1, companies earning 8% or more beat profitable companies earning less by 17.2 percentage points over the 21 months. Holding size, sector and dividends constant barely changes that gap: it stays at 16.9 points. Paying a dividend, by contrast, added just 2.1 points, too small to tell apart from chance.

What dividends do

Dividends do one thing well: they cushion a weak market.

We tested seven periods between January 2025 and October 2026, from six to 21 months long. In three of them, the typical SGX share price fell or barely moved, ranging from a 6.5% fall to a 2.3% rise. One of these took in 7 April 2025, when the Straits Times Index fell 7.46% in a single day. In those three periods, dividend payers’ share prices did 7.2 to 10.6 percentage points better than companies of similar size, sector and return that paid nothing.

In the other four periods, the typical share price rose between 9.7% and 18.1%. There, paying a dividend made no measurable difference.

Dividends protect shareholders when markets fall. They do not lift valuations, which is what the market reforms led by MAS and SGX set out to do.

Disclosure cannot do the market’s job

The regulators have been candid about the limits of disclosure. “By themselves, these activities will not move a stock,” Mr Tan said in April. “That is not what regulation is for. That is the job of the market.” In March, Chia Der Jiun, managing director of the Monetary Authority of Singapore, warned that mandatory disclosure on its own “carries the risk of being a box ticking exercise, if there is no real ownership by company leadership.”

A recent column in The Edge Singapore, on why investors pass over small and mid-sized companies, was headlined “Apathy, not ignorance, hurting Singapore’s SMID counters”. Our data suggests one reason for that apathy. Investors can already see which companies earn more than their cost of equity. A dividend policy tells them nothing new.

RegCo had reasons to keep the dividend rule loose. Boards need room to change course, and a fixed policy could tie their hands. A company may also depart from its policy, as long as it explains why.

If disclosure cannot move a stock, the fair test of a rule is whether it changes what boards do. By that test, the dividend rule asks the least.

RegCo does ask that a policy be “clearly articulated” and give “meaningful insight”. Yet its own sample shows how little that requires. The consultation paper offers two versions. In the first, the company fills in a payout ratio of its own choosing. The second commits to nothing: the board “will declare or recommend dividends where it considers that doing so is appropriate.” RegCo offers both as examples of how a policy may be described.

Boards already serious about returns will write a good policy; those that are not will write a vague one. Investors pay for what a company earns, not for what its policy says.

The pay rule asks more. RegCo set a standard a company can fail: “A generic statement that remuneration is determined by reference to financial and non-financial measures, without identifying the KPIs, would not satisfy the rule.” It also reaches what managers are paid to deliver, which shapes the return a company earns. Even so, it asks companies to describe their measures, not to change them. And neither rule says what happens to a company that falls short.

Big on support, light on returns

Japan and Korea have tried to lift valuations a different way (See Table 1). Both put the same question at the centre of what they asked companies: does the company earn more than its cost of capital?

In March 2023, Japan’s stock exchange asked listed companies to act on their cost of capital and share price, and it publishes each month the names of those that have responded. In July 2023 it also launched the JPX Prime 150, an index of large companies whose return on equity exceeds their cost of equity and which trade above the value of their net assets. Korea added tax breaks for companies that return more to shareholders and, from July 2025, a legal duty for directors to act for shareholders as well as the company.

Singapore’s package is broad and well funded. MAS has committed S$6.5 billion to fund managers to invest in Singapore shares, and pays for more research coverage of listed companies. Its S$30 million Value Unlock programme, running from 2026 to 2028, works at the level of the individual company. The Equip grant pays half the cost of training, up to S$15,000 a company. The Elevate grant pays half the cost of outside advisers on strategy, financial management and investor relations, up to S$200,000 a company. Companies taking it must publish value plans and performance targets.

What Singapore does not do is ask every company the question Japan and Korea put first. Its new compulsory rules cover dividends, pay and investor relations. None of them, and none of the grants, asks whether a company earns more than its cost of capital. The closest the package comes is the review of the Code of Corporate Governance, which MAS says will consider guidance on how boards challenge management on “capital efficiency”.

Pay managers for returns above the cost of equity

Boards do not need a rule to act. The surest route to a higher share price is to earn more than the cost of equity, year after year. To keep management focused on that, boards can build it into long-term pay, measured over three to five years. What gets measured gets managed.

RegCo has pointed the way. Its consultation paper lists return on equity and return on invested capital as examples of pay measures, because they show “whether value is created above the cost of capital”. Boards can go one step further and make that return a pass-or-fail test. Managers may hit every other target, but if the company has not earned more than its cost of capital, the long-term award does not pay out.

SATS already works this way. Its long-term awards are measured on shareholder returns, profit and carbon targets, but are “further contingent upon meeting a performance threshold, which requires Return on Invested Capital to exceed the Weighted Average Cost of Capital in order for vesting to occur.” SGX bases its variable pay pool on “profits above the cost of equity hurdle”.

There is a cost. Managers may cut long-term investment to hit a short-term number. Measuring over three to five years deals with most of that, provided the remuneration committee holds the bar when results fall short.

Many SGX companies have a controlling shareholder, whose interests in dividends, pay and the use of capital do not always match those of other shareholders. The board’s task is to bring them into line, and to hold itself to the same measure it sets for management.

Mr Tan said investors would judge “whether this is the best use of cash on the balance sheet.” They already have. Companies earning above their cost of equity trade at 1.63 times their net assets; profitable companies earning less, at 0.59 times. A dividend policy will not change that verdict. Better returns will.

Written by Lee Ooi Keong, 24 October 2026

Lee Ooi Keong is an Independent Director of an SGX Mainboard-listed company with 30 years of experience in corporate performance, investments and risk management. He is the founder and Managing Director of Clover Point Consultants, an independent Board and C-suite advisory firm, and was formerly Director of Risk Management at Temasek for over 16 years.