Your ownership percentage can fall while the value of your holding stays the same. It can also fall while you become wealthier—or while value is transferred away from you. The word dilution identifies a change in the claim. It does not, by itself, complete the economic diagnosis.
The missing question is what the company received in exchange for the additional shares. New ownership claims might bring cash, a productive business, employee services or the removal of a debt obligation. They might also be issued on terms that are expensive for existing owners. Counting the new shares is necessary. Following the exchange is equally necessary.
This guide separates ownership percentage, voting power, earnings per share and economic value. Its worked cases show why those quantities sometimes move together and sometimes do not. All examples are fictional; each calculation uses explicitly simplified assumptions rather than a forecast of an actual share price.
The article continues the equity-ownership series within How Finance Works. Securities Issuance explains the capital-raising process. This page concentrates on the before-and-after position of the shareholder whose claim is being changed.
Educational scope: general ownership and valuation mechanics, with identified Singapore and US reference points. Rights, conversion terms, accounting treatment and deadlines depend on the instrument and jurisdiction. This is not a recommendation to subscribe, sell, borrow or invest.
What share dilution means
In its simplest ownership sense, dilution occurs when the denominator of relevant ownership claims increases faster than a particular holder’s own participation. For one class of economically identical shares, the holder’s percentage is their shares divided by all outstanding shares in that class.
If you hold 10 million of 100 million equal shares, you own 10%. If the company issues another 25 million equal shares and you buy none, your holding becomes 10 million out of 125 million, or 8%. Your account can still display exactly the same 10 million shares. The change is in the denominator.
Singapore’s MoneySense explains dilution in the context of rights issues, where existing shareholders are offered an opportunity to buy new shares proportionately. The SEC also describes dilution arising when convertible securities become ordinary shares. See MoneySense’s shares guide and Investor.gov on convertible securities.
For navigation, begin with the four dilution questions, then a fairly priced issue, an economically discounted issue, rights issues, option-pool denominators, and a practical verification method.
Four different questions hide inside one word
An ownership percentage measures a fraction of defined claims. Voting power measures a fraction of eligible votes. EPS divides an earnings amount by an accounting share denominator. Economic value asks what the holder’s claim is worth after the transaction. Those quantities need separate labels.
| Question | Quantity being tested | What must be checked |
|---|---|---|
| Do I own a smaller fraction? | Ownership percentage | Holder shares and the relevant total count. |
| Can I influence less of the vote? | Voting power | Votes attached to each class and the voting denominator. |
| Does each share represent less reported earnings? | Earnings per share | The earnings allocation and the appropriate weighted share count. |
| Is my holding worth less? | Economic value | Value received, rights transferred, costs and future operating consequences. |
The first question can often be answered with a capitalisation table. The last requires a valuation argument. A percentage calculation cannot substitute for that argument, and an optimistic valuation cannot erase an actual loss of voting influence.
This separation is the central discipline of the article. A transaction can be dilutive under one measure, neutral under another and beneficial under a third without any mathematical contradiction.
Percentage points and percentage reductions are not the same
In the opening example, ownership fell from 10% to 8%. That is a decline of two percentage points. Relative to the original 10% position, it is a 20% reduction: two divided by ten.
The new investors hold 25 million out of 125 million shares, or 20% of the post-issue company. Existing shareholders collectively retain 80%. Each non-participating existing holder keeps 80% of their former ownership percentage when all shares have the same relevant rights.
Calling the change “2% dilution” without identifying the basis is ambiguous. It might mean two percentage points of total ownership or a 2% relative reduction in the holder’s position. A useful calculation writes both the before-and-after percentages and the relative change.
A smaller percentage without a loss of value
Consider fictional Harbour Systems. It has 100 million identical ordinary shares and an estimated equity value of S$800 million before a funding round. The estimated value is therefore S$8 per share. Lina holds 10 million shares, representing 10% and an estimated S$80 million.
The company issues 25 million new equal shares at S$8 each. It receives S$200 million. Assume there are no fees, no change in the existing business’s value, no special rights, and that the new cash adds its full amount to equity value. These are teaching assumptions, not a prediction of how a market will react.
| Before issue | After issue | |
|---|---|---|
| Estimated total equity value | S$800 million | S$1,000 million |
| Total equal shares | 100 million | 125 million |
| Estimated value per share | S$8 | S$8 |
| Lina’s shares | 10 million | 10 million |
| Lina’s ownership | 10% | 8% |
| Estimated value of Lina’s holding | S$80 million | S$80 million |
Lina owns a smaller fraction of a larger total. Her percentage is diluted, but the model does not reduce the value of her holding. The new investors paid for the additional claim at the assumed fair pre-issue price.
This does not prove the financing is desirable. It establishes a narrower point: percentage dilution alone is insufficient evidence of an economic loss. The actual judgement depends on the price, rights, costs and use of capital.
When the issue price transfers value
Keep the same pre-issue estimate of S$800 million and 100 million shares, but change the offering. The company now issues 25 million equal shares at S$4 each, receiving only S$100 million. Assume again that cash adds dollar for dollar and that nothing else changes.
The post-issue equity estimate is S$900 million. Dividing by 125 million shares gives S$7.20 per share. Lina’s 10 million shares are now estimated at S$72 million rather than S$80 million. Her percentage falls to the same 8% as before, but this time her modelled value also falls.
Across all existing shareholders, the original 100 million shares are now worth S$720 million under the model: an S$80 million reduction. The new investors’ 25 million shares are worth S$180 million after paying S$100 million. The S$80 million difference makes the transfer visible.
The result depends on the assumed S$800 million pre-issue economic value. An issue below yesterday’s quotation does not automatically establish an unfair economic discount. Yesterday’s price might not describe current prospects, transaction risk or available funding alternatives. The model isolates the effect of issuing below a stated value estimate; it does not prove that estimate for any real company.
The useful distinction is therefore between a discount to a reference market price and a discount to a defensible estimate of economic value. The existing Price vs Value guide explains why those are not interchangeable.
Follow the net contribution, not only the headline proceeds
Return to the S$200 million issue at S$8, but assume S$10 million of transaction costs reduce the new resource available to the business. Holding all other teaching assumptions unchanged, post-issue equity value becomes S$990 million. Dividing by 125 million shares gives S$7.92.
The example shows why gross capital raised is not the same as net value received. A share issue can be priced at an assumed fair value and still involve real costs. Those costs do not disappear because the transaction successfully closes.
A general analytical bridge is: begin with the pre-transaction equity estimate, add the net resources received, then incorporate any incremental change in business value that has not already been counted. Divide by the relevant post-transaction claims, respecting differences between share classes.
The phrase “not already been counted” matters. If a project valuation already includes the cost of deploying the new cash, adding the project’s gross value and leaving the same cash untouched can count one resource twice. The companion Market Capitalisation vs Enterprise Value develops this boundary discipline.
Dilution can finance an improvement—or only postpone a problem
Consider two possible uses of the S$200 million. In one scenario, the money funds a capability whose estimated net present value is positive after its investment cost. In another, the money repeatedly covers operating losses without repairing the cause. The same issue size and share count can lead to different economic futures.
This is why the analysis should follow the capital through time. What expenditure happens first? What operating change is expected? When does cash return? What evidence would show the plan is working? What happens if it does not?
The point is not that a company must avoid loss-making investment periods. Some valuable projects need a period of spending before their benefits arrive. The question is whether the funding buys a credible route to value or merely another interval before a similar financing request.
Use Financial Runway to examine how much decision time the financing buys, and Capital Allocation to examine what the company does with that time and money.
An EPS-accretive acquisition can still reduce value per share
Use another fictional case. The buyer has 100 million equal shares, estimated equity value of S$800 million and ordinary earnings of S$80 million. Its estimated value is S$8 per share and its EPS is S$0.80.
It acquires a target by issuing 25 million new shares. Assume the target adds S$180 million of economic equity value and S$25 million of comparable annual earnings, with no synergies, transaction costs, additional financing effects or special rights. Treat the combined earnings and shares as present for a full teaching period.
Combined estimated equity value is S$980 million across 125 million shares, or S$7.84 per share. Combined earnings are S$105 million, producing S$0.84 EPS. EPS rises 5%, while estimated value per share falls 2%.
The two outcomes are possible because earnings and economic value are different quantities. The share exchange can increase the current earnings attached to each share while giving away too much ownership relative to the value acquired under the model.
The example does not establish that a particular acquisition destroys value. It demonstrates why “EPS accretive” is not a complete investment appraisal. The target’s risk, reinvestment needs, future growth and price paid still matter. Read Earnings per Share and Mergers and Acquisitions for the neighbouring mechanisms.
Rights issues: an opportunity to participate, not free new ownership
A rights issue offers existing shareholders the opportunity to buy newly created shares in a stated proportion within a specified period. MoneySense explains both the subscription opportunity and the dilution that can follow non-participation. The offer still requires an assessment of what the company plans to do with the money. See MoneySense on rights issues.
Imagine a one-for-four offer: one new share at S$6 for every four existing shares. Assume the existing share price is S$10 before the rights separate, all shares have equal rights, the offer is fully subscribed, and there are no costs or changes in business value apart from the cash received.
Four old shares represent S$40. The new subscription contributes S$6. The five post-issue shares therefore represent S$46 in this theoretical model, giving a theoretical ex-rights price of S$9.20.
The S$6 subscription price should not be compared with S$10 as though the new shares appear without changing the denominator. Both the cash and the shares must be added before the post-issue position is interpreted.
Three rights-issue outcomes for the same holder
Suppose Lina owns 100 old shares worth S$1,000 under the preceding assumptions. Her proportional entitlement allows her to subscribe for 25 new shares.
She subscribes in full. She pays S$150 and holds 125 shares. At the theoretical S$9.20 price, the position is worth S$1,150. Subtracting her additional S$150 contribution leaves the same S$1,000 of original value. She preserves her ownership proportion in the fully subscribed issue, but does so by committing more money.
She sells a tradable rights entitlement. Under an idealised renounceable-rights model, the aggregate theoretical value of the rights associated with her 100 old shares is S$80. Her old shares become worth S$920 and selling the rights for S$80 would preserve the S$1,000 total before fees and other effects. Her ownership percentage still falls; the theoretical compensation is separate from retaining the percentage.
She allows the entitlement to expire unused. Her 100 shares are worth S$920 at the model price, and the S$80 rights value has not been realised. That is a different outcome from selling the rights. Doing nothing should not be treated as automatically equivalent to a costless decision.
Not all rights are tradable, and actual prices need not match the model. DBS’s guide for its CPFIS and SRS customers distinguishes subscription, sale where permitted and lapse, while directing customers to the particular event’s terms and deadlines. It is an account-specific illustration, not a universal dealing procedure. See DBS’s rights-events guide.
Theoretical ex-rights value is not a price guarantee
The S$9.20 calculation deliberately holds the business outlook still while adding cash and shares. A real funding announcement can change how investors judge the business. The amount raised, its intended use, the need for emergency funding and changes in control can all become part of the valuation question.
The teaching model therefore answers a conditional question: what follows arithmetically if the stated value assumptions remain unchanged? It does not promise that the shares or rights will trade at the theoretical number.
The distinction keeps the model useful. It provides a baseline against which an actual difference can be investigated, rather than a prediction that replaces the investigation.
Option pools: define the denominator before discussing the percentage
Suppose a fictional company has 100 million issued ordinary shares. A proposed financing model includes a reserve for 20 million possible employee shares in its fully diluted denominator. Under that definition, the reserve represents 20 divided by 120, or approximately 16.67% of the modelled total—not 20%.
To make the reserve 20% of the post-reserve total, solve x ÷ (100 + x) = 20%. The answer is 25 million. The modelled denominator becomes 125 million and the reserve is one-fifth of it.
This is a denominator problem, not a dispute about whether employee incentives are worthwhile. “A 20% pool” is incomplete until the agreement specifies 20% of which total, at which moment, with which other securities included.
For this example, the reserve is a modelling assumption about potential shares, not a statement that ungranted shares have already become legally outstanding. A real capitalisation table must distinguish issued shares, outstanding awards, ungranted reserves and conditional conversion rights. The term fully diluted must be read according to the transaction’s own definition.
Employee shares exchange ownership for something else
Imagine that an employee receives shares in exchange for services that improve the company’s capacity. The analysis cannot stop with “more shares means worse.” It must ask what contribution was obtained, what alternative compensation would have cost and how much ownership was transferred.
Nor should it stop with “no cash was paid, therefore the service was free.” Existing owners can surrender economic participation even when the company does not write a cash cheque for the entire compensation amount.
In an original before-and-after model, compare the estimated business outcome with and without the service and the associated award. Hold the time horizon and risks consistent. That exercise may support the award or challenge it; the percentage alone cannot do either job.
The accounting expense, legal share issuance and eventual economic return also need their own timelines. One event can be recognised, granted, vested and settled at different stages. The correct treatment of a real award requires its terms rather than a generic assumption about all employee equity.
Reported diluted EPS is not a complete future ownership scenario
IAS 33 applies specific tests to potential ordinary shares and excludes antidilutive effects from diluted EPS. An exclusion from that accounting denominator does not cancel the underlying instrument. See the IFRS Foundation’s IAS 33 overview and the detailed measurement requirements.
This creates an important reading distinction. A loss-making company’s basic and diluted EPS may be identical while potential future shares remain relevant to ownership. An accounting metric describing one period should not be used as proof that no future denominator change is possible.
The companion EPS guide owns the reporting calculation. Here, the task is to build a separate future scenario with explicit exercise, conversion and issuance assumptions.
Conversion can change the share count without bringing fresh cash that day
A convertible security can become ordinary shares under its contractual formula. Investor.gov distinguishes fixed conversion arrangements from market-price-based formulas and explains why falling prices can increase the number of shares issued under some variable structures. See the SEC’s convertible-securities explanation.
For arithmetic only, assume S$10 million of a claim converts at S$2 per share. It produces 5 million shares. If a different contractual scenario sets the conversion price at S$1, the same amount produces 10 million shares. Floors, caps, accrued amounts and other terms could change a real calculation.
The share-count effect is only one side. If the company received the money when the security was originally issued, it should not count the same cash again at conversion. The conversion instead changes the claim structure, potentially removing a debt or preference claim while creating ordinary ownership.
A coherent before-and-after map therefore follows shares, liabilities, interest or distribution obligations and any actual cash movement together. Otherwise a model can exaggerate the cost by ignoring a claim that disappeared—or exaggerate the benefit by inventing new cash that never arrived.
Repeated rounds multiply the remaining ownership
Suppose Lina starts with 12% and does not participate in three later rounds. Under the example’s equal-share assumptions, existing holders retain 80% of the post-round company in the first round, 75% in the second and 90% in the third.
Her final ownership is 12% × 80% × 75% × 90% = 6.48%. She keeps 54% of her original percentage, a 46% relative reduction. The rounds cannot be evaluated by adding percentage-point labels without defining each denominator.
Now ask about value. A 6.48% interest in a much stronger company can be worth more than 12% of the original company. It can also be worth less. The answer requires the resources raised, the rights issued and the business outcomes—not just the multiplication.
The useful capitalisation history therefore records both the fraction retained and the value-producing reason for each round. A sequence of smaller percentages is a history of ownership changes, not automatically a history of either success or failure.
Voting dilution can differ from economic dilution
Different share classes can carry different voting rights. MoneySense’s dual-class guide explains how votes per share can separate control from economic ownership. See MoneySense on dual-class shares.
In a fictional issue of non-voting shares with equal economic participation, the economic denominator might increase without adding ordinary votes. In another structure, a small number of high-vote shares might change control much more than the headline share count suggests.
The calculation must therefore specify the right being divided. One fully diluted economic table is not automatically a voting-power table. The first article in this series, Shares and Shareholder Rights, explains that distinction in greater depth.
A split is not dilution merely because the share count rises
A proportional stock split increases everyone’s share units together and does not, by itself, reduce their ownership percentage. Investor.gov explicitly distinguishes the split from ownership dilution. See the SEC’s stock-split definition.
If Lina owns 100 shares out of 10,000, she has 1%. After a two-for-one split she owns 200 out of 20,000: still 1%. A theoretical halving of price leaves the same position value before any separate market change.
The distinction is mathematical. New units allocated proportionately are not the same event as new claims allocated only to someone else in exchange for resources. Counting units without tracking who receives them produces the wrong conclusion.
Buybacks can offset share issuance without making it free
Suppose a fictional company begins with 100 million outstanding shares, issues 4 million new shares and repurchases 4 million shares that cease to be outstanding. Its ending count is again 100 million.
If the repurchase price is S$10 per share, the company has spent S$40 million to remove those shares. The stable ending count does not make that resource use disappear. It also does not establish the annual EPS denominator, because the transaction dates can affect the weighted average.
The right interpretation records the gross issuance, the repurchase, the cash paid and the resource received for the issuance. A net share-count change of zero is one observation, not the complete economic history.
Do not double-count costs when integrating this with a valuation. A compensation cost, a cash repurchase and a change in claims must each be placed in the model consistently. The existing Share Buybacks guide owns the broader allocation decision.
Compare a financing with the relevant alternative
A proposed issue can be highly dilutive and still be better for existing holders than the feasible alternative. To see the logic without prescribing an action, imagine a company whose old equity is estimated at almost nothing if it cannot obtain funding. A restructuring that leaves old holders with a smaller stake in a viable business may improve their position relative to that unfunded scenario.
That comparison does not justify any price or any transfer of control. It says the baseline matters. Comparing a rescue issue with an imaginary world in which the company has abundant costless funding produces a misleading verdict.
The same discipline can challenge an unnecessary issue. If a company could meet its needs through a less costly alternative, surrendering additional ownership may be difficult to justify. The analytical task is to compare feasible scenarios on consistent assumptions, not to classify all dilution as either clever or harmful.
A practical verification method
Begin with a dated before-and-after capitalisation table. Identify the actual outstanding shares, each class’s economic and voting rights, the proposed new claims and any conditional instruments. Mark a reserve, grant, exercise, conversion and completed issuance as different events rather than combining them under one vague share total.
Next, trace the consideration. What cash, assets, services or released obligations does the company receive? What fees or additional commitments accompany the exchange? Has a payment arrived now, or was it received in an earlier period? This step prevents both missing value and invented value.
Then calculate the holder’s position using the appropriate denominator. State percentage points separately from percentage reductions. Build economic value as a separate estimate, show its assumptions and reconcile it back to the relevant share classes.
Finally, identify the operational follow-through. Does the holder need to respond to a corporate action? Which institution holds the shares? Which documents specify eligibility and deadlines? ACRA’s overview of share transactions helps distinguish transaction types, while the actual issuer and intermediary documents govern the particular event.
This is a proposed reading workflow, not a substitute for professional advice. Its purpose is to make the reasoning reproducible: another reader should be able to see which numbers are reported facts, which are contractual assumptions and which are valuation judgements.
An observable mastery test
A company issues 25% more equal ordinary shares. A shareholder who does not participate keeps the same number of shares but sees their percentage fall. Management says the transaction increases EPS. An online comment says that the shareholder has automatically lost 25% of their money.
A careful answer begins with the denominator. Issuing 25% more shares means existing holders collectively retain 100 divided by 125, or 80%, of their former ownership fractions: a 20% relative reduction, not 25%. Whether their money value fell requires the price, resources received and resulting business value.
The EPS statement also needs its own bridge. What earnings were added, what costs changed and what share count was used? The acquisition example showed that higher EPS can coexist with lower estimated value per share.
You understand dilution when you can replace both confident slogans with a complete calculation: the claim before, the exchange, the claim after and the assumptions needed to connect ownership to value.
The return to the real business
Funding or operating need → new claims → resources received → changed ownership and control → capital use → future earnings and cash → value reaching each holder.
The share count records how participation is divided. The business determines what that participation can ultimately be worth. Good analysis keeps both visible.
A smaller fraction is not automatically a smaller fortune. A successful capital raise is not automatically a good exchange. The decisive question is what existing owners surrendered, what the company obtained and whether the resulting business can justify the new claims.
Sources and further reading
The external reference points are MoneySense’s guides to shares and rights issues and dual-class structures, the SEC’s explanations of convertible securities and stock splits, the IFRS Foundation’s IAS 33 overview, ACRA’s share-transaction overview, and DBS’s account-specific rights-events guidance. The numerical cases are original illustrations; theoretical prices are not promised market outcomes.
Continue through the equity-ownership series
Read Shares and Shareholder Rights to identify the claim, Market Capitalisation vs Enterprise Value to identify the value boundary, and Earnings per Share to read the earnings denominator. Return to How Finance Works for the complete financial-system map.