Distributions: being paid for owning

A dividend is a decision, not a debt

A company's board declares a dividend. Nothing compels it to. A record of fifty consecutive annual increases is a reputational commitment that boards defend hard — and it is still not a contract, which is why long records end in recessions.

Preferred shares sit between a bond and a common share. They carry a stated rate and must be paid before any common dividend, and a cumulative preferred accrues anything skipped. But skipping is generally permitted: missing a preferred dividend does not put the issuer in default the way missing a bond coupon does. That single difference explains most of the gap in what the two pay.

Funds add a third case. A closed-end fund or ETF may operate a stated distribution policy — a fixed monthly or quarterly amount. That is a promise about the schedule and the size, not about the source, and the difference between those two things is the subject of this lesson.

The four dates

Four dates govern every distribution. On the declaration date the payer announces it. On the ex-dividend date the shares begin trading without the right to it. The record date identifies the holders on the books. The payable date is when cash actually arrives, often weeks later.

The ex-date is the one that decides entitlement: buy on or after it and the seller keeps the payment. There is no window in which you can buy a dividend cheaply, because on the ex-date the share price is adjusted down by roughly the amount being paid. The cash is leaving the company, and the market prices the company after it leaves.

This is also why payment frequency is a scheduling fact rather than an income fact. A monthly payer and a quarterly payer distributing the same annual amount pay the same amount; the monthly one is simply smoother, and reinvests a little sooner.

Payout ratio, and where it means nothing

For an ordinary operating company, the payout ratio — dividends divided by earnings — is a rough measure of room. A ratio drifting up while earnings sit flat means the dividend is claiming a growing share of a fixed pie, and the cushion is thinning.

For a REIT the same ratio is close to meaningless. Accounting rules require large depreciation charges on buildings that in many cases are not losing value, so reported earnings sit far below the cash the properties actually generate. Payout ratios above 100% of earnings are routine and tell you nothing. The measures built for the job are funds from operations and adjusted funds from operations, which add depreciation back and, in the adjusted version, subtract the recurring capital spending a building really needs.

Every pass-through structure has its own denominator. For a business development company it is net investment income — the interest it collects less its costs — and whether distributions are covered by that, or are leaning on realised gains or on capital. For a master limited partnership it is distributable cash flow and the coverage ratio built from it. For a fund there is no earnings figure at all: a fund distributes what it receives, what it realises, and whatever else it has decided to pay.

So 'is the payout ratio safe' is really the question 'which denominator belongs to this structure', and using the wrong one produces confident nonsense in both directions.

Why REITs, BDCs and MLPs pay out the way they do

These structures exist to avoid a layer of tax. A REIT that meets the statutory tests, including distributing at least 90% of its taxable income to shareholders, is not taxed at the entity level on what it distributes. A business development company that qualifies as a regulated investment company faces a similar distribution requirement. High payouts are therefore a structural obligation, not generosity.

The obligation has a consequence that is easy to miss. A company distributing nearly all of its taxable income retains almost nothing to grow with, so expansion has to be funded by issuing new shares or by borrowing. That makes these structures unusually sensitive to their own share price and to credit conditions: when raising capital is expensive, growth stops, and the distribution has to carry the whole return.

Master limited partnerships take a different route to the same place. A partnership is not taxed as an entity; income is allocated to the partners and reported on a Schedule K-1 rather than a 1099. Distributions frequently exceed the income allocated to you, and the excess reduces your cost basis rather than being taxed now — which is pleasant during ownership and shows up as a larger taxable gain at sale.

Not all of a distribution is income: return of capital

Part of what a fund or partnership pays you can be classified as a return of capital: your own money coming back. It is not taxed in the year you receive it, and it reduces your cost basis, so the tax arrives later as a larger gain or a smaller loss.

There are two very different reasons for it. Sometimes return of capital is an artefact of tax accounting — a REIT's depreciation, or a partnership's allocations — and the underlying business is producing every dollar it pays. Sometimes it means a fund is distributing more than it earns and is shrinking itself to keep the schedule intact. Both appear as the same line on a statement.

The way to tell them apart is to look, not to guess. The 1099-DIV separates non-dividend distributions. Closed-end funds paying under a managed distribution policy issue Section 19(a) notices estimating how much of each payment is income, gains and capital, with the final character determined after year-end. Net asset value over several years, read alongside the distribution history, tells the rest of the story.

Taxes on distributions

Qualified dividends from US corporations, held long enough, are taxed at long-term capital gains rates. Ordinary dividends are taxed at your marginal rate. Which one you have is determined by the payer and the holding period, not by your intentions.

Pass-through structures are mostly ordinary. REIT distributions are largely ordinary income, with portions sometimes characterised as capital gain or return of capital; BDC distributions are largely ordinary. MLPs bring a K-1, state filing obligations where the partnership operates, and unrelated business taxable income issues that make them awkward inside retirement accounts.

Foreign payers add withholding at source, reduced by treaty in many cases and sometimes recoverable as a foreign tax credit — a real cost that headline yields never show. In all of these, the character of what you received is decided by the payer's accounting and reported to you after the year has ended.

What makes a distribution stop

Distributions stop for reasons that are visible before the announcement. Coverage erodes: the earnings, funds from operations or net investment income behind the payment shrink toward it and then below it. Leverage tightens: a covenant restricts payments, or refinancing at a higher rate absorbs the cash the distribution was coming from. A floating-rate preferred resets to a lower rate on its scheduled date. A fund exhausts the gains it was distributing.

The cut itself is the last event in that sequence, not the first. Which is why the research question is never 'what is the yield' but 'what is behind the payment, and what is happening to it'.

None of this predicts any particular cut, and nothing here says a given security is safe or unsafe. The point is to know which number to watch for the structure in front of you.