Dividend investing isn't about yield—it's about finding which companies actually return cash
Dividends are cash and can't deceive; the author uses his real trade of selling Citigroup at $51 before the financial crisis to show that high dividend yields are often illusions from price crashes, not signals the company is improving.
The video won't play here. Listen to the audio instead:
The argument · tap a timestamp to hear it
AI capital spending is draining free cash flow; dividend stocks become an alternative
The host asks why dividends matter in an AI- and semiconductor-dominated market. Bonsall counters that this isn't dividend versus total return, but a way to achieve total return. He notes that historically the most profitable free-cash-flow companies now run negative free cash flows due to high capex, making companies that return profits to shareholders and maintain healthy balance sheets carry different risk characteristics. He cites Broadcom and Blackstone—AI-related companies that are themselves excellent dividend-growth stocks—showing dividend investing and AI narratives are not mutually exclusive.
— David BonsallDividends are cash; EBITDA can be massaged
Bonsall's core logic: tax forms can understate income but no one overstates it because you don't pay tax on money you didn't make. Dividends follow the same principle—they're actual cash payouts, while metrics like EBITDA leave ample room for unverifiable adjustments. A company willing to sustain and grow dividends signals management's genuine confidence in business growth; slowing dividend growth may reflect management's true doubts. He argues cash hitting the bank account beats whatever press releases and 10-Q filings claim.
— David BonsallBuybacks are often executive compensation, not capital return to shareholders
On large-scale buybacks like Apple's, Bonsall offers a counterintuitive observation: if the company is truly buying back stock aggressively, why are shares outstanding higher than before the buybacks started? Net share issuance should be negative, but in reality Silicon Valley companies use buybacks mainly to offset dilution from executive compensation, not to genuinely return capital to shareholders. He's not against buybacks per se, but notes that investors living on pensions or cash flow ‘can't eat buybacks’, and when trouble hits, buybacks are cut first while dividends—protected by decades-old ‘social contract’—are last to fall. His funds' portfolios saw zero dividend cuts through the 2008 financial crisis and the pandemic.
— David BonsallStock halving doesn't make dividends better—this is the classic yield trap
Using arithmetic, Bonsall exposes the illusion of ‘high dividend yield’: at $100 per share with 4% yield you collect $4; when the stock falls to $50 you still collect $4, but your stated yield jumps to 8% while you've actually lost 50% of your capital—nothing got better. He calls such stocks ‘accidental high-yielders’ and stresses that what you want is dividend yield rising because the per-share payout grows, not fake yield created by stock collapse. This distinction is the line between ‘high-yield stocks’ and ‘dividend-growth stocks’.
— David BonsallSelling Citigroup at $51: identifying dividend traps requires qualitative judgment
Bonsall recounts a career-defining trade: in October 2007, after Citigroup announced it would not cut dividends, he sold at $51—the stock later fell to 97 cents. The reason: Citigroup simultaneously disclosed $10.5 billion in bonds at roughly 10% interest from Saudi and Kuwaiti investors to sustain the dividend. ‘Borrowing on credit cards to pay bills is not a sustainable way to fund dividends.’ His warning signs: when a company stops raising its dividend after 20 years of growth—even without cutting—or when growth slows, these are qualitative red flags demanding scrutiny beyond quantitative metrics and fundamental research.
— David BonsallDividend yield itself is a valuation ruler, not just P/E
Bonsall describes a valuation method they actually use for buy-sell decisions: benchmark a company against its own historical dividend yield. Walmart's historical average dividend yield is about 3.5%, but after its stock surged 60% in a year, the yield fell to 1.5%—near the S&P 500 average—and despite pristine fundamentals they sold their lifetime Walmart position. Broadcom is another case: stock up 16-fold in a year; though it raised dividends 14% annually for eight consecutive years, the yield compressed to 1%, triggering a sell decision.
— David BonsallDividend-growth stocks have 0.6 beta; inflation-beating is their natural attribute
Bonsall notes dividend-growth portfolios run about 0.6 beta to the broad market, thus naturally lower market sensitivity—in 2008 only two Dow stocks rose: Walmart and McDonald's, both top dividend-growth stocks over the prior 50 years. More crucially, dividend growth historically beat inflation in every calendar year: if your portfolio grows dividends 6% to 9% while inflation runs 2% to 5%, you're already winning purely on cash-flow grounds. But he quantifies opportunity cost: from 2000 to 2009 the S&P saw zero return yet dividend-growth portfolios averaged 7-8% annually; from 2009 to 2026's bull market the S&P averaged 15% but dividend-growth portfolios only beat the index in about half the years.
— David BonsallThirty stocks provide adequate diversification; cost-basis returns over twenty years hit 50%-100%
On building a dividend-growth portfolio, Bonsall offers specific rules: hold 30 to 35 companies, cap each position at 1.5% to 4%, small enough not to matter but large enough to matter for diversification. He stresses that fewer than 80 companies in the S&P 500 now offer consistent dividend growth, showing dividend culture itself is shifting; seeking concentration and certainty while still diversified requires care. On sell discipline: beyond ‘bad reasons’ like deteriorating fundamentals (Citigroup, BP's oil spill), the more common trigger is ‘good reasons’—stocks like Walmart, Apollo, and Broadcom run up so much their yields compress to 1%, forcing switches to 3% or 4% names. That's opportunity-cost-driven rebalancing.
— David BonsallIn their own words · checked verbatim
Well, the dividends don't lie, they're cash.
David Bonsall2:06
share buybacks are executive compensation. They are not return of capital to shareholders.
David Bonsall6:19
You can't eat a stock buyback.
David Bonsall7:22
Borrowing from your credit card to pay your bills is not how a dividend gets sustained
David Bonsall9:26
why is dividend yield itself not a valuation metric, its own historical dividend yield?
David Bonsall12:30
You can't fool the fundamentals.
David Bonsall18:41
Figures
| Citigroup share price sold, Oct 2007 | $51 | 9:26 |
| Citigroup stock price fell to | $0.97 | 9:26 |
| Broadcom dividend growth, 8 consecutive years | 14% | 13:33 |
| Dividend-growth portfolio beta | ~0.6 | 14:36 |
| S&P 500 cumulative return, 2000-2009 | 0% | 15:39 |
| S&P 500 annualized return, 2009-2026 | 15% | 15:39 |
| Holdings per portfolio | 30-35 | 20:48 |
| Single position weight | 1.5%-4% | 20:48 |
| McDonald's total dividend reinvestment return | 400,000% | 18:41 |
Glossary
- Yield trap
- Stock price decline that inflates the dividend yield on paper, misleading investors into thinking dividends have improved.
- Free cash flow
- Cash generated from operations minus capital expenditures—the cash a company can freely deploy.
- Payout ratio
- The percentage of corporate profits paid out as dividends.
- Net share issuance
- New shares issued minus shares repurchased; the net change in shares outstanding.
How to listen
Individual investors focused on defensive asset allocation and worried about AI valuation bubbles; investment advisers wanting to understand the stock-picking logic of dividend-growth strategies.