Dividends Are Not Income: A Quantitative Reality CheckS&P 500SP:SPXEdgeToolsDividend investing is one of the most popular strategies among. The idea is straightforward: buy stocks that pay regular dividends, and you receive a steady stream of income. Entire YouTube channels and subreddits are built around it. But there is a problem with this framework. Not a matter of opinion, but of accounting. Dividends are not income in the way most investors understand the word. And once you see the data on what this misunderstanding actually costs, you can make better decisions. That is what this study is for. We test every common argument for dividend investing against 20 years of market data, and where the data finds a weakness, we show what works better. 1. How dividends actually work When a company pays a $1 dividend, its stock price drops by exactly $1 on the ex-dividend date. A $100 stock paying $1 becomes a $99 stock plus $1 in cash. Your total wealth has not changed. The company transferred part of its equity to you as cash. This is not a theoretical detail. It is how exchanges handle every single dividend payment. Modigliani and Miller formalized this in 1961: in the absence of frictions, dividend policy is irrelevant to the value of a firm. An investor who wants cash can sell shares. An investor who wants growth can reinvest dividends. The company's decision to pay or not pay dividends does not change the underlying business. In practice, one friction matters enormously: taxes. Dividends are taxed when received. Capital gains on share price appreciation are taxed only at the point of sale. This timing difference compounds over years and decades. The rest of this study quantifies exactly how much. 2. The data We compare the S&P 500 (SPY) against seven dedicated dividend ETFs, all using total return data (dividends reinvested) from Tiingo: SPY (S&P 500): 1993 to 2026, 33 years VIG (Vanguard Dividend Appreciation): 2006 to 2026, 20 years VYM (Vanguard High Dividend Yield): 2006 to 2026, 19 years SCHD (Schwab US Dividend Equity): 2011 to 2026, 15 years NOBL (ProShares S&P 500 Dividend Aristocrats): 2013 to 2026, 13 years DVY (iShares Select Dividend): 2003 to 2026, 23 years SDY (SPDR S&P Dividend): 2005 to 2026, 21 years HDV (iShares Core High Dividend): 2011 to 2026, 15 years A methodological note: these ETFs have different inception dates. SCHD launched in 2011, after the 2008 financial crisis. NOBL started in 2013. This creates a bias because ETFs that missed a major crash appear to have better risk metrics than they would have earned through the full period. We address this throughout the study by showing two views for every comparison: each ETF from its own start date, and all ETFs from the same common start date (October 2013). Fig. 1: Growth of $1 invested. Upper panel: SPY, VIG, VYM, DVY, and SDY from November 2006, including the 2008 crisis. Lower panel: all eight ETFs from October 2013. Log scale. Figure 01 provides the first overview. SPY outperforms the dividend ETFs in both the long-history view and the common period view. This does not mean dividend ETFs are bad investments in absolute terms. Several of them produced solid positive returns. The question is whether the dividend focus adds value compared to simply holding the broad market. 3. The dividend component of returns Fig. 2: Upper panel: total return (solid) vs. price-only return (dashed) for SPY, VIG, VYM, and SCHD. Lower panel: the dividend gap as a percentage of total return, measured from the common start date October 2011. Figure 02 separates each ETF's total return into two components: price appreciation and dividends. The gap between the solid and dashed lines represents the cumulative contribution of dividends. Over the common period since 2011, SPY's dividend gap grew to about 35%, VYM's to 27%, VIG's to 25%, and SCHD's to 23%. SPY shows the largest gap despite paying the lowest yield because its stronger total return produces a larger absolute dividend contribution. This decomposition matters because the two components are taxed differently. Price appreciation is tax-deferred until you sell, giving you control over when to realize gains. Dividends are taxed in the year you receive them, with no choice about timing. The higher the dividend share of your returns, the more of your wealth is subject to annual taxation. 4. Downside behavior Fig. 3: Drawdowns from prior peak. Upper panel: each ETF from inception. Lower panel: all ETFs from October 2013. A common argument for dividend stocks is that they offer better downside protection. The data is mixed. During the 2008 crisis, DVY fell 62.6% and VYM fell 57.0%, both deeper than SPY's 55.2%. The companies with the highest dividend yields in 2007 were concentrated in Financials and Energy, exactly the sectors hit hardest. In contrast, VIG (-46.8%) held up somewhat better because it screens for dividend growth rather than high yield, which tends to select for more stable businesses. The lower panel of Figure 03 compares all ETFs from October 2013 onward. During the COVID crash in 2020, DVY dropped about 42% while SPY fell 34%. SCHD and NOBL were roughly in line with SPY. Bear-market Sharpe ratios confirm the inconsistency. The Sharpe ratio measures return per unit of risk taken. In bear markets, all values are negative, but less negative is better: SPY: -0.426 VIG: -0.315 VYM: -0.385 SCHD: -0.584 NOBL: -0.451 DVY: -0.460 SDY: -0.199 HDV: -0.436 VIG and SDY held up better than SPY. DVY, SCHD, NOBL, and HDV did worse. If your thesis for owning dividend stocks is crash protection, the data shows it depends entirely on which dividend ETF you pick and which crash you experience. It is not a systematic advantage. 5. Performance comparison Fig. 4: Six performance metrics in two blocks. Top block: each ETF since inception. Bottom block: all ETFs from October 2013. Fig. 5: Calendar year returns for all ETFs, 2006 to 2026. Empty cells indicate periods before the ETF existed. The annualized returns and Sharpe ratios for each ETF since inception: SPY: 10.76% return, 0.579 Sharpe, -55.2% max drawdown VIG: 10.06% return, 0.590 Sharpe, -46.8% max drawdown VYM: 9.30% return, 0.504 Sharpe, -57.0% max drawdown SCHD: 13.24% return, 0.857 Sharpe, -33.4% max drawdown NOBL: 10.31% return, 0.649 Sharpe, -35.4% max drawdown DVY: 8.94% return, 0.477 Sharpe, -62.6% max drawdown SDY: 8.94% return, 0.470 Sharpe, -54.8% max drawdown HDV: 10.62% return, 0.728 Sharpe, -37.0% max drawdown VYM, DVY, and SDY trail SPY in both return and risk-adjusted terms. Over the 19-year VYM period, $100,000 in VYM grew to $562,000 while the same amount in SPY reached $737,000. That is a $175,000 difference. SCHD is the strongest dividend ETF with 13.24% annualized. Two important qualifications: its history begins in October 2011, after the 2008 crisis, as the empty cells in Figure 05 show. And when measured over the common period from October 2013, SPY leads with 14.4% annualized and a Sharpe of 0.84, while SCHD comes in at 12.4% and 0.78. SCHD is a good fund, but its headline numbers benefit from starting at a favorable point in market history. 6. Yield, persistence, and correlation Fig. 6: Trailing 12-month dividend yield. DVY and SDY run at 3-5%, SPY below 2%. The 2008 spike reflects falling prices rather than rising dividends. Fig. 7: Rolling excess return of each dividend ETF vs. SPY over 1, 3, and 5-year windows. Fig. 8: Daily return correlation matrix. Most dividend ETFs correlate above 0.92 with each other. DVY and SDY yield 3-5%, roughly triple what SPY pays. Higher yield means more cash distributed, but also a larger annual tax burden. Figure 07 shows an important pattern: the outperformance of dividend ETFs over SPY is not persistent. Rolling 5-year excess returns were briefly positive around 2011-2013, then turned negative for most dividend ETFs from 2018 onward as technology-driven growth dominated the market. Investors who chose dividend ETFs expecting sustained outperformance did not find it. Figure 08 addresses whether holding multiple dividend ETFs provides diversification. It does not. Most dividend ETFs correlate above 0.92 with each other. VYM and SCHD correlate at 0.98. The underlying stock selections are different, but the return characteristics are nearly identical. 7. What dividend ETFs actually are Fig. 9: Left: R-squared from factor regression. All above 0.89. Right: annualized alpha with t-statistics. No ETF reaches significance at t > 2.0. This section contains the study's central finding. We regressed each dividend ETF's daily returns against six factor proxies: MTUM (Momentum), VLUE (Value), QUAL (Quality), IWD (Russell 1000 Value), IWF (Russell 1000 Growth), and USMV (Minimum Volatility). A factor regression asks a simple question: how much of this ETF's behavior can be explained by known market tilts? If there is a genuine "dividend premium," we would expect to find some leftover return (alpha) that the factors cannot explain. The R-squared values range from 0.89 to 0.97. Known factors explain 89 to 97 percent of the variation in dividend ETF returns. And the alpha is not statistically significant for any of them. Statistical significance requires a t-statistic above 2.0; the highest any dividend ETF achieves is SCHD at 1.7. What the regressions reveal about each ETF: VYM loads primarily on the Value factor (IWD beta = 0.694, t = 41.3). It is essentially a value fund. VIG loads on Quality (QUAL beta = 0.269) and Low Volatility (USMV beta = 0.390). It is a quality/low-vol fund. SCHD combines Value, Quality, and Low Volatility. Its strong returns are explained by these factor tilts, not by anything specific to dividends. This is a constructive finding. The qualities that make dividend-paying companies attractive (strong cash flows, stable business models, reasonable valuations) are real. They are just not unique to dividends. They are the Value and Quality factors, studied for decades and available through dedicated factor ETFs like QUAL, VLUE, or USMV. These factor ETFs give you the same underlying exposure without the forced annual tax event that dividends create. 8. The cost of forced distributions Fig. 11: Tax drag simulation for Germany (26.375%), USA (15%), and Switzerland (35%). Starting capital $100,000, 19.4-year period. We simulated two investors over the 19.4-year common period of SPY and VYM, each starting with $100,000. Investor A holds SPY. Capital gains tax is deferred until sale. Final value: $736,702. Investor B holds VYM, pays German tax on dividends (26.375%), and reinvests the remainder. Final value: $453,508. The difference is $283,194 or 38.4% of the accumulating portfolio's terminal value. This gap has two components: SPY's higher total return and the tax drag from VYM's larger dividend yield. Both work in the same direction. At other tax rates: US qualified dividends (15%): VYM reaches $497,483 (gap: $239,219) Swiss income tax (35% marginal): VYM reaches $422,741 (gap: $313,961) A note on how these rates work in practice, because the headline numbers do not tell the full story: In Switzerland, the often-cited 35% Verrechnungssteuer (anticipatory tax) on Swiss-source dividends is not a final tax. It is a withholding designed to encourage proper tax reporting. Swiss residents who declare their securities and dividend income correctly in their tax return receive the full 35% back, either as a direct refund or as a credit against cantonal and municipal taxes. The actual tax on dividends equals the investor's personal marginal income tax rate. For US-listed ETFs like SPY or VYM held by a Zurich resident, the applicable rates are: 15% US withholding tax (reduced from 30% under the US-Switzerland double taxation agreement), plus Swiss income tax on the gross dividend, minus a credit for the US tax already paid. The effective combined rate depends on the investor's marginal rate. Our simulation uses 35% as a marginal income tax rate, not the Verrechnungssteuer. In Germany, the 26.375% Abgeltungssteuer (25% plus 5.5% solidarity surcharge) is withheld automatically by the broker. However, the first EUR 1,000 of capital income per year (EUR 2,000 for married couples filing jointly) is exempt through the Freistellungsauftrag (exemption order), which must be filed with the broker. Investors whose personal income tax rate is below 25% can reclaim the difference through the Guenstigerpruefung (favorability check) in their tax return. This applies mainly to students, part-time workers, and retirees with low income. For most working investors, the 26.375% is the effective final rate. In the United States, qualified dividends are taxed at 0%, 15%, or 20% depending on taxable income. The 0% rate applies to single filers with income up to approximately $48,350 and married couples up to $96,700 (2025 brackets). This means many retirees and lower-income investors pay no federal tax on qualified dividends at all. Our simulation uses the 15% rate, which applies to the majority of working investors. High earners above $200,000 (single) may also owe an additional 3.8% Net Investment Income Tax. These mechanisms do not eliminate the fundamental tax timing disadvantage of dividends (immediate taxation vs. deferred capital gains), but they reduce the effective rates in specific circumstances. Investors who qualify for reduced rates face a smaller tax drag than our simulation shows. 9. The alternative: homemade dividends Fig. 12: Homemade dividend (SPY, selling 3% per year) vs. real dividend (VYM), at the German tax rate. If the goal is regular cash flow from a portfolio, there is a straightforward alternative. Instead of buying dividend stocks and receiving forced distributions, buy a broad market fund and sell a small percentage each year. This approach, sometimes called a "homemade dividend," was proposed by Modigliani and Miller in 1961 and has a practical advantage: you control when and how much to sell, which gives you control over your tax timing. We tested this directly. Two investors both want 3% annual cash flow, starting with $100,000. Investor A buys VYM, collects and reinvests dividends, withdraws 3% per year. After 19 years: Remaining portfolio: $248,000. Total cash withdrawn: $71,000. Combined wealth: $320,000. Investor B buys SPY, sells 3% per year. After 19 years: Remaining portfolio: $411,000. Total cash withdrawn: $88,000. Combined wealth: $500,000. The SPY investor ends with $180,000 more in combined wealth and received $17,000 more in actual cash withdrawals. The homemade dividend produced more income and more terminal wealth simultaneously. 10. Sector exposure and market regimes Fig. 18: Sector betas from regressing ETF returns against all 11 SPDR sector ETFs. Fig. 10: Sharpe ratios across six market regimes. Fig. 13: Monthly return distribution for all ETFs. Fig. 14: Maximum drawdown and recovery time. Upper panel: since inception. Lower panel: common period from October 2013. One dimension often overlooked: dividend ETFs carry implicit sector bets. We regressed each ETF's returns against all 11 SPDR sector ETFs (XLK, XLF, XLV, XLY, XLP, XLE, XLI, XLU, XLRE, XLC, XLB). SPY's largest exposure is Technology, which makes sense given tech's weight in the S&P 500. VYM's top loadings are Financials, Consumer Staples, and Industrials. Technology ranks seventh. SCHD is similar: Consumer Staples and Industrials dominate. This is worth understanding because it reframes the performance comparison. Dividend ETFs have not underperformed because dividends are inherently bad. They have underperformed because their sector composition underweights Technology, the market's strongest sector from 2010 to 2025. Whether this continues is an open question, but every investor in dividend ETFs should be aware they are expressing a sector view, whether they intended to or not. On market regimes: a common claim is that dividend stocks outperform when interest rates fall because they act as "bond substitutes." The data does not support this clearly. All ETFs in the study had higher Sharpe ratios when rates were rising than when they were falling. There is no consistent regime advantage for dividend strategies. 11. International markets Fig. 15: International dividend and broad-market ETFs from their common start date. One qualification to the US-focused findings: internationally, dividend ETFs have performed better. VYMI (Vanguard International High Dividend Yield) achieved a higher Sharpe ratio than broad international alternatives VEA and VWO over the available 2016-2026 period. The sample is short at 10 years and untested through a major international crisis, so this finding should be treated as preliminary rather than conclusive. 12. Long-term compounding effects Fig. 16: Theoretical 30-year portfolio growth at 8% annual return with 3% dividend yield under different tax rates. For investors with multi-decade horizons, the compounding effect of annual dividend taxes becomes substantial. With 8% growth and a 3% dividend yield over 30 years, starting with $100,000: Accumulating (tax deferred): $1,006,266 US tax at 15%: $864,629 German tax at 26.375%: $778,743 Swiss tax at 35%: $727,609 The German investor loses $227,523 to taxes that could have been deferred. The loss accelerates over time: roughly $4,000 in year 5, over $15,000 by year 30. Tax drag compounds in the same way that returns compound, just in the opposite direction. 13. Where dividend investing does work The findings of this study do not apply equally in all contexts. Tax-advantaged accounts: In an IRA, 401(k), Roth IRA, or similar tax-sheltered structure, dividend taxes do not apply during the holding period. The tax drag that makes dividend investing expensive in taxable accounts disappears entirely. In these accounts, dividend ETFs can be a perfectly reasonable choice based on their factor characteristics. Behavioral considerations: Some investors find it psychologically easier to spend dividends than to sell shares. If the alternative to a dividend portfolio is an accumulating portfolio that never gets rebalanced or withdrawn from, the dividend approach may produce better real-world outcomes despite its tax inefficiency. Specific tax jurisdictions: Some countries tax capital gains more heavily than dividends. In those cases, the analysis reverses. The constructive takeaway: the qualities that attract investors to dividend stocks (stable businesses, disciplined capital allocation, reasonable valuations) are real and worth pursuing. They correspond to the well-documented Value and Quality factors. Investors who want these characteristics can access them through factor ETFs like QUAL, VLUE, or USMV while retaining control over the timing of their tax events. 14. Limitations The ETF histories are short. SCHD starts in 2011, NOBL in 2013. Neither has been tested through a major financial crisis. Their risk metrics may be optimistic. The tax simulation uses fixed rates throughout. Real rates change and individual circumstances vary. The dollar amounts illustrate the mechanism; they are not predictions. The factor regression uses ETF proxies (MTUM, VLUE, IWD), not the academic Fama-French factor series. Results may differ slightly with academic data, though the direction would not change. Individual stock dividend strategies (Dogs of the Dow, manual dividend aristocrat selection) were not tested because survivorship bias in pre-2010 data cannot be corrected with our sources. Figures 01, 02, 03, 04, and 14 each include a common-period panel to control for inception-date bias at the ETF level. Fig. 17: Summary dashboard combining equity curves, Sharpe ratios, max drawdowns, factor R-squared, and tax simulation. 15. Summary Dividends are a return of capital, not income. They trigger a tax event that capital gains do not. Across 7 dividend ETFs, 19 years, and 3 tax jurisdictions, the data shows: 1. Most dividend ETFs trail the S&P 500 in total return 2. All of them are explained by known factors (Value, Quality, Low Volatility). There is no unique dividend alpha. 3. Downside protection is inconsistent across ETFs and market events 4. The tax drag costs up to 38% of terminal wealth over 19 years at German rates 5. A homemade dividend strategy using SPY outperforms VYM dividends by $180,000 on a $100,000 portfolio For investors in taxable accounts, the evidence points toward broad market or factor ETFs with a systematic withdrawal plan. For investors in tax-sheltered accounts, dividend ETFs remain a reasonable option. In either case, understanding what dividends actually are (a return of your own equity, not new income) is the foundation for making better portfolio decisions. References Arnott, R.D., Hsu, J.C. and Moore, P. (2005) 'Fundamental Indexation', Financial Analysts Journal, 61(2), pp. 83-99. Baker, M., Nagel, S. and Wurgler, J. (2007) 'The Effect of Dividends on Consumption', Brookings Papers on Economic Activity, 2007(1), pp. 231-291. Black, F. (1976) 'The Dividend Puzzle', Journal of Portfolio Management, 2(2), pp. 5-8. Fama, E.F. and French, K.R. (1993) 'Common Risk Factors in the Returns on Stocks and Bonds', Journal of Financial Economics, 33(1), pp. 3-56. Fama, E.F. and French, K.R. (2001) 'Disappearing Dividends: Changing Firm Characteristics or Lower Propensity to Pay?', Journal of Financial Economics, 60(1), pp. 3-43. Hartzmark, S.M. and Solomon, D.H. (2019) 'The Dividend Disconnect', Journal of Finance, 74(5), pp. 2153-2199. Modigliani, F. and Miller, M.H. (1961) 'Dividend Policy, Growth, and the Valuation of Shares', Journal of Business, 34(4), pp. 411-433. Shefrin, H.M. and Statman, M. (1984) 'Explaining Investor Preference for Cash Dividends', Journal of Financial Economics, 13(2), pp. 253-282. Data sources Price data: Tiingo API (adjusted close for total return, unadjusted close for yield calculations), with TwelveData and AlphaVantage as fallback providers Dividend data: Financial Modeling Prep API, AlphaVantage API Macro data: FRED (Federal Reserve Economic Data) Factor ETF proxies: MTUM, VLUE, QUAL, IWD, IWF, USMV via Tiingo Sector ETF proxies: XLK, XLF, XLV, XLY, XLP, XLE, XLI, XLU, XLRE, XLC, XLB via Tiingo Tax rates: Germany 26.375% (KapESt + Soli), USA 15% (qualified), Switzerland 35% (marginal) All data and code available at EdgeTools/research/29_dividend_vs_buyhold/ Disclaimer This study is for educational and informational purposes only. Past performance does not guarantee future results. This is not investment advice. Individual tax situations vary. Consult a qualified tax advisor for personal financial decisions.