Exploring Radical Views on Dividend Policy

Two of the biggest names in dividend theory, John Lintner and Myron Gordon, argued that shareholders actively prefer a bird in hand: a rupee paid out today over an uncertain rupee promised tomorrow. Franco Modigliani and Merton Miller disagreed entirely, showing that in a perfect market, dividend policy has no bearing on a firm’s value at all. Between these two positions sits a third, messier camp of thinking that doesn’t try to settle the relevance-versus-irrelevance argument. Instead, it asks a more practical question: what do dividends actually do inside a real company, with real managers, real taxes, and real information gaps? These are the radical views on dividend policy, and they explain far more about how Indian boardrooms actually behave than the classical models do.

Why the radical views emerged 🔗

The Modigliani-Miller irrelevance proposition rests on a set of assumptions that rarely hold outside a textbook: no taxes, no transaction costs, no information gap between managers and shareholders, and no conflict of interest inside the firm. Once you relax even one of these assumptions, dividend policy starts to matter again, just not for the reasons Gordon and Lintner proposed. The radical school picks apart these frictions one at a time. It looks at what happens when managers know more than shareholders, when insiders have incentives that don’t perfectly match owners’ interests, and when different investors face different tax bills on dividends versus capital gains. Three ideas dominate this body of work: agency cost theory, signaling theory, and tax preference or clientele theory. Each one treats the dividend not as a value-neutral transfer of cash, but as a tool that does real work inside the corporate structure.

The agency cost view: dividends as a check on managers 🔗

Once ownership and management are separated, as they are in almost every large public company, a gap opens up between what shareholders want and what managers actually do. Managers control cash, and cash creates temptation: pet projects, oversized offices, unnecessary acquisitions, or simply a comfortable cushion that reduces the pressure to perform. This is the agency problem, first framed rigorously by Jensen and Meckling, and dividend policy turns out to be one of the cheapest tools available to control it.

Frank Easterbrook’s argument, built on this idea, is elegantly simple. If a firm pays out a large share of its earnings as dividends, it depletes its internal cash reserves and is forced to return to the capital markets whenever it needs money for new projects. That return trip means external scrutiny: investment bankers, credit rating agencies, and institutional investors all get a chance to question the firm’s plans before financing is approved. In this way, dividends work as a kind of recurring audit on management, lowering the agency costs that would otherwise build up around a cash-rich, loosely monitored management team, a mechanism explored in detail in the literature on dividend payout determinants. Later research applying agency and signaling theory together has found that larger, more mature firms with fewer growth avenues tend to pay out more precisely because monitoring their reinvestment choices matters more once easy growth options run out, a pattern documented in recent work on dividend policy and financial constraints.

India’s cash-hoarding problem and the buyback wave 🔗

This is not an abstract debate in India. A 2019 study by governance advisory firm IiAS found that BSE500 companies were together sitting on a cash pile worth several trillion rupees, with dozens of firms able to comfortably return a large share of it to shareholders without hurting operations, as reported by Business Standard’s coverage of the report. IT majors were named repeatedly. TCS and Infosys, both generating far more free cash flow than their business needed, faced sustained pressure from investors and activist shareholders to either invest the surplus productively or hand it back. TCS responded with what was then the largest share buyback in Indian corporate history, a decision that followed pointed investor commentary on the need for a clearer, more disciplined payout policy. Commentators at the time noted that idle cash on a company’s books typically earns a far lower return than the same money could generate if deployed in the core business or handed to shareholders to reinvest elsewhere, a point laid out clearly in this analysis of India’s cash-hoarding problem. That is agency cost theory playing out in real time: cash sitting unused on a balance sheet is precisely the situation the theory predicts will invite investor pressure for higher payouts.

The signaling theory: dividends as a language investors read 🔗

Managers almost always know more about a company’s true earning power than the shareholders who own it. This information gap is the starting point for signaling theory. A dividend increase is expensive and hard to reverse without embarrassment, which is exactly what makes it a credible signal. When a company raises its dividend, it is effectively telling the market that it expects future cash flows to comfortably support the higher payout, and investors tend to take that seriously. A survey of managers at consistently dividend-paying firms found that signaling was, by a clear margin, the explanation managers themselves gave most credit to when asked why their firms paid dividends at all, well ahead of tax-based or agency-based reasoning, according to research on managerial perspectives on dividend policy.

The flip side matters just as much. A dividend cut is read as bad news almost automatically, because managers are known to protect dividend payments for as long as they possibly can. When a cut does happen, the market usually assumes the situation is worse than what the company is willing to say outright, and share prices tend to react sharply and immediately.

Cracks in the signaling story 🔗

Signaling theory has an uncomfortable internal contradiction. Dividends are heavily concentrated among large, established, profitable companies, which are precisely the firms that suffer the least from information asymmetry in the first place. If signaling were the whole story, younger and less-understood firms with more to prove would be the ones paying the biggest dividends, not the other way round. There is also evidence of a slow, multi-year drift in stock returns following dividend announcements rather than an instant, clean price adjustment, which sits awkwardly with the idea that dividends are efficient, one-shot signals, a critique detailed in this working paper on dividend policy and controlling shareholders. Signaling theory explains a lot, but not everything.

Tax preference and the clientele effect 🔗

Dividends and capital gains have rarely been taxed identically, and that difference shapes what shareholders actually want from a company. Where capital gains are taxed at a lower rate or only when shares are sold, and dividends are taxed immediately on receipt, rational investors in higher tax brackets have an incentive to prefer companies that reinvest profits and grow share value rather than companies that pay out cash every year. This is the core of tax preference theory, laid out clearly in this overview of academic dividend policy models.

India’s own tax history illustrates the point well. Dividend Distribution Tax, once paid by the company itself, was scrapped in 2020, and dividend income is now taxed directly in the hands of the shareholder at their applicable slab rate. For an investor in the highest tax bracket, that can make dividend income noticeably less attractive after tax than an equivalent gain locked in through a lower long-term capital gains rate. Not every investor faces the same calculation, though, which is where the clientele effect comes in. Retired investors and income-focused funds often want steady cash payouts regardless of the tax hit, because they need the income to live on or to meet regular payout obligations. Younger, high-earning investors in growth-oriented portfolios frequently prefer the company to reinvest instead. Firms respond by settling into a payout policy that matches the investor base they want to attract and then largely sticking with it, since a sudden change in policy risks driving away the very clientele the firm has built.

The other side of the coin: does paying dividends starve growth? 🔗

Every one of these radical arguments explains why dividends can be useful. Critics point out that usefulness has a cost. Every rupee paid out as a dividend is a rupee that cannot be reinvested internally, and internal funds are almost always the cheapest source of capital a company has. External financing, whether debt or fresh equity, comes with issuance costs, covenants, and often a less favourable price than management believes the company deserves, precisely because outside investors face the same information gap that makes signaling theory work in the first place. A firm that commits to a high, sticky dividend policy may end up starving genuinely profitable projects of funding, or forced to raise expensive external capital for them instead.

This tension shows up clearly among Indian IT companies today. Several large IT firms have kept dividends and buybacks flowing even as revenue growth has slowed sharply, partly because their businesses are mature and cash-generative, and partly because investors have come to expect it. Analysts tracking the sector have flagged that sustaining big payouts without matching earnings growth can strain reserves over time, even for companies with historically strong cash positions, as discussed in recent reporting on IT sector dividend and buyback trends. The same payout that agency theory would call disciplined can, from a growth investor’s chair, look like a company that has quietly run out of ideas.

None of these three views fully displaces the others, and none of them settles the old Gordon-versus-Modigliani-Miller debate either. What they do is shift the question from whether dividends affect share value in theory to what dividends actually accomplish inside a real firm with real conflicts of interest, real information gaps, and real tax bills. That shift is probably why finance professionals lean on these ideas far more than on the cleaner classical models when they sit down to actually set a payout policy.

What do you think? When a cash-rich company like TCS or Infosys announces a large buyback, do you read it as disciplined capital allocation that agency theory would approve of, or as a quieter signal that the company has run short of growth opportunities worth funding? And if you were building an investment portfolio today, would you deliberately seek out a dividend clientele that matches your own tax situation and income needs, or does that kind of fine-tuning matter less than it sounds?

References

1. https://www.westga.edu/~bquest/1999/payout.html
2. https://pmc.ncbi.nlm.nih.gov/articles/PMC10569952/
3. https://www.business-standard.com/amp/article/markets/india-inc-s-achilles-heel-bse500-firms-are-sitting-on-a-huge-cash-pile-119041800032_1.html
4. https://www.business-standard.com/amp/article/opinion/amit-tandon-a-way-out-of-the-cash-hoarding-problem-117022600726_1.html
5. https://link.springer.com/article/10.1007/BF02759711
6. https://www.ecgi.global/sites/default/files/working_papers/documents/SSRN-id2412588.pdf
7. https://pressbooks.pub/tourocorporatefinance/chapter/7-4-dividend-policy-theories-2/
8. https://upstox.com/news/market-news/stocks/tcs-to-infosys-it-companies-balance-weak-growth-with-dividends-and-buybacks/article-183361/