DarkRange55
We are now gods but for the wisdom
- Oct 15, 2023
- 2,059
When financial historians survey two centuries of investment history across dozens of countries, a remarkably consistent pattern emerges. Assets fall into a hierarchy of long-run performance. Equities dominate as the unrivaled compounding engine, turning modest annual returns into dynastic wealth over generations. Bonds occupy a middle position, delivering modest but generally positive real growth while protecting against some risks. Bills and cash-like instruments offer liquidity and nominal stability but fail to preserve value once inflation is considered. Gold, though romanticized as a timeless store of value, belongs in a different category. Its role is not to generate wealth, but to preserve purchasing power across centuries, buffering against the destruction of fiat regimes and crises without offering meaningful compounding.
The differences between these assets appear small when expressed as annual returns, but over decades and centuries, they diverge dramatically. Equities returning 6–7% real per year double purchasing power roughly every decade, multiplying a single dollar into more than a thousand across a century. Bonds, at 1–2% real, may turn that dollar into twenty or thirty. Bills and deposits, hovering close to zero after inflation, barely preserve capital at all. Gold, meanwhile, stands apart. Its purchasing power has shown uncanny stability, not growth. The same ounce of gold that could outfit a Roman centurion with a fine cloak and sandals, or buy a gentleman's suit in 1890s America, still buys a fine suit today. Its power is cultural continuity, not compounding.
Roy Jastram's classic The Golden Constant (1977, updated 2009 by Jill Leyland) captured this phenomenon with centuries of data from England and the United States. His results showed that gold's purchasing power mean-reverts. In inflationary times, gold lags in the short run but eventually catches up as fiat currencies devalue. In deflation, gold's purchasing power rises as consumer prices fall faster than gold's nominal value. Across the long arc of history, the real return of gold averages essentially zero. It is reliable for continuity, not growth.
More recent research confirms this. Campbell Harvey, Claude Erb, and John Welsh extended Jastram's work to 2020, analyzing gold back to 1257. They concluded that gold has returned about 0.5–1% annually in real terms, versus ~7% for U.S. equities. Robert Barro and Sanjay Misra studied gold from 1836 to 2011 and reached a similar conclusion: a 1.1% real return, dwarfed by equities but better than cash. The evidence is overwhelming: gold preserves, equities multiply.
The story of equities is one of extraordinary compounding. Jeremy Siegel's Stocks for the Long Rundocuments that U.S. equities have averaged ~6.7% real CAGR since 1802. Dimson, Marsh, and Staunton put the global figure at ~5.3% since 1900, with the U.S. achieving ~8.5%. Jorion and Goetzmann's study of 39 countries since 1921 found equities averaged ~4.3% real, but warned of survivorship bias since many markets were wiped out by wars or expropriation. The key point is that equities dominate where they endure.
The mathematics of compounding makes this dominance unavoidable. A dollar invested in U.S. equities in 1900, with dividends reinvested, would be worth more than $2,000 in real terms today. The same dollar in gold would scarcely exceed $10. Equities' superiority is structural, tied to dividends, productivity, and retained earnings. Gold may safeguard wealth, but only equities multiply it.
Dividends are central to equities' supremacy. Studies by the CFA Institute and others show that the majority of long-term equity wealth arises from reinvested dividends rather than price appreciation. Gold, in contrast, generates no cash flow. Its scarcity supports its role as a hedge, but its inertness limits its long-run performance.
This divide is the essence of the equity premium puzzle. Rajnish Mehra and Edward Prescott showed in 1985 that equities outperformed Treasury bills by ~6% annually over the twentieth century, far beyond what risk models could justify. Later work by Campbell and Shiller, and Fama and French, confirmed the persistence of this anomaly. Equities are consistently rewarded with extraordinary returns, while gold, behaving like an inert store of value, returns little beyond inflation.
Other studies reinforced this. Santa-Clara and Valkanov, Keim and Stambaugh, and the UBS Yearbooks all confirmed that equities offer a persistent, global premium. Survivorship bias explains part of this — many firms fail — but the index as a whole reflects the productivity of surviving enterprises. Gold avoids bankruptcy but offers no growth; equities are risky individually but explosive in aggregate.
Bonds occupy the middle rung. Sidney Homer and Richard Sylla's A History of Interest Ratesdocuments thousands of years of fixed-income markets, while modern data from DMS, Barclays, and Deutsche Bank show government bonds earning ~1–2% real CAGR since 1900. Robert Shiller's U.S. data back to 1871 confirm this figure. Bonds produce positive real growth but are vulnerable to inflation.
This vulnerability is clear in Antti Ilmanen's Expected Returns. Bonds thrive in disinflationary periods such as 1980–2000, when falling yields boosted prices. In inflationary decades like the 1970s, bonds were crushed in real terms, while gold soared. Over the long run, bonds provide modest compounding, while gold provides continuity.
Bills and CDs are weaker still. Data from Ibbotson and DMS show bills averaging ~0.5–0.8% real since 1900. They serve as liquid, nominally safe assets but are eroded by inflation. John Campbell and Luis Viceira argued they are useful for short-term stability but cannot preserve multigenerational wealth. Gold, despite lacking yield, performs better than cash across centuries because fiat currencies erode.
Volatility comparisons further illuminate the trade-offs. Equities and gold have both shown ~15–20% annualized volatility since 1971. Yet equities delivered ~7% real CAGR, while gold delivered close to zero. McCown and Zimmerman described gold as a "zero-beta asset" — moving independently of equities, but still volatile. Investors often misinterpret gold's safe-haven role as implying stability, but in practice gold can be just as wild as stocks.
Indeed, equities experience dramatic crashes but tend to rebound. The Great Depression erased nearly 80% of U.S. equity value, and 2008 cut markets in half. Yet dividends and earnings growth restored wealth. Gold suffers different risks: long grinding drawdowns. Between 1980 and 1999, gold lost ~70% nominally and ~80% in real terms, taking more than three decades to recover.
And gold's volatility can even surpass that of equities. Standard deviation during crisis decades has spiked above 25–30%, higher than most equity markets. Its movements are driven less by fundamentals and more by macro shocks: inflation surges, currency collapses, or changes in the monetary system. Erb and Harvey showed that gold's volatility clustered in crisis decades, with extreme swings in the 1970s, the long collapse of the 1980s–1990s, and the rebound during 2008–2011.
This profile creates an asymmetrical burden. In calm decades such as the 1980s and 1990s, equities provided rising returns while gold bled value. In crisis decades, gold's gains came with chaotic swings rivaling equities. Deutsche Bank confirmed that gold's drawdowns are often longer and its recoveries entirely dependent on rare systemic shocks. Analysts like Claude Erb stress that gold is not "low risk" but "different risk": useful for tail events, punishing in stability.
Correlation evidence helps explain why investors still allocate to gold. Dirk Baur and Brian Lucey found gold acts as a hedge on average and a safe haven in crises. Baur and McDermott extended this internationally, confirming the pattern. Virginie Coudert and Hélène Raymond quantified the effect: during U.S. recessions from 1978 to 2009, gold gained ~12% while equities lost ~4%. In bear markets, gold's losses averaged just 1%, compared to equities' 20%.
Survivorship bias in equities underscores the difference. Hendrik Bessembinder found in 2018 that just 4% of stocks accounted for all net wealth creation in the U.S. from 1926 to 2016. The vast majority matched T-bills or destroyed wealth. Gold, though unproductive, cannot fail outright. It lacks upside but avoids the risk of total ruin that haunts individual equities.
One of the clearest ways to illustrate gold's weakness as a long-term investment is through its Sharpe ratio, the standard measure of risk-adjusted returns. Since the collapse of Bretton Woods in 1971, gold has returned about 6–7% nominal per year but with annualized volatility of roughly 15–20%. Adjusted for inflation, its real return is closer to 1–2%, producing a Sharpe ratio near zero over long horizons. By contrast, U.S. equities in the same period have delivered real returns of 6–7% with similar volatility, giving them a Sharpe ratio closer to 0.3–0.4, depending on the sample window (Dimson, Marsh, Staunton; UBS Yearbook 2023). Bonds, though lower-returning, often achieve Sharpe ratios in the 0.2–0.3 range because of reduced volatility. In other words, every unit of risk taken in gold has historically produced little or no incremental reward, confirming Claude Erb and Campbell Harvey's conclusion that gold's primary utility lies not in its risk-adjusted performance but in its diversification properties during crises.
Gold's drawdown history is one of the strongest reminders that volatility does not mean guaranteed growth. After peaking in 1980 near $850/oz, gold spent almost two decades in decline. By 1999 it had fallen to about $250/oz, representing a 70% nominal loss and closer to 80% in real terms once inflation was considered. This slump lasted longer than any single equity bear market in modern history, and only by 2011 did gold finally reclaim and surpass its 1980 highs. Equities, by contrast, while suffering devastating collapses in the Great Depression (nearly –80%) and again in 2008 (–50%), recovered through earnings growth and dividends. Gold has no such recovery mechanism; it must wait for the next crisis or inflationary surge to find support.
This contrast is central to the psychology of investors. Equities are tied to the productive capacity of firms. As long as profits return, equity markets recover, and those who reinvest dividends are rewarded. Gold, however, offers no organic growth. Its long periods of stagnation, like the 1980–1999 slump, create opportunity costs that can devastate long-term savers if it is treated as a primary investment. This is why scholars like Claude Erb and Campbell Harvey describe gold not as a growth asset but as "different risk," one that hedges specific scenarios but often fails in normal conditions.
Yet bonds provide a counterpoint, offering moderate but genuine compounding. Long-term government bonds have delivered around 2% real annual returns since 1900, according to the Dimson–Marsh–Staunton database and Robert Shiller's U.S. series. But bonds' strength depends heavily on the regime. In the 1970s, when inflation averaged over 7% per year, bondholders suffered losses exceeding 40% in real terms, while gold rose more than 400%. In the disinflationary 1980–2000 era, however, bonds produced their best century-long run, as yields fell from double digits to below 5%, while gold collapsed. Antti Ilmanen's Expected Returns illustrates this "mirror image": gold thrives when inflation erodes fixed-income assets, but bonds thrive when inflation is tamed.
Historical episodes show this dynamic clearly. During the U.S. Civil War, the government issued "greenbacks," paper notes not fully backed by gold. Bonds tied to these notes plummeted in value as inflation surged, while gold became the black-market medium of choice for merchants. Homer and Sylla's History of Interest Rates notes that gold premiums in New York reached nearly 200% at the war's peak. In both World Wars, government bonds were subject to yield caps and inflation, meaning their real returns were negative. Gold, though formally pegged, was hoarded in private markets and held its value more effectively. These episodes confirm that gold serves best as a hedge against extreme fiscal stress, while bonds are reliable in stability.
Bills and short-term deposits occupy a still weaker position. Ibbotson's SBBI Yearbook and DMS data show Treasury bills averaging only 0.5–0.8% real returns since 1900. In calm times they serve their purpose — providing liquidity, stability, and a nominally safe asset. But in inflationary or wartime settings, they are destroyed in real terms. During the Great Depression, Treasury bills preserved nominal value but produced near-zero real return. Gold, though fixed in U.S. law at $20.67/oz, was revalued by Franklin Roosevelt in 1933 to $35, a 69% jump that handed holders a windfall while bills sat stagnant.
The World War II era illustrates this divergence again. Treasury yields were held artificially low by government policy, often below 2%. Inflation, however, ran higher, wiping out real gains. Gold, though officially fixed, retained underground value and provided continuity of wealth across borders. In practice, bills eroded while gold endured. Modern certificates of deposit tell a similar story. In the early 1980s, nominal CD rates touched 15%, but inflation was nearly as high, leaving real returns modest. Over the long run, cash-like instruments lag every other asset class, beaten even by inert gold.
The volatility comparison sharpens when extended globally. Japan's "lost decades" offer an instructive case. After peaking in 1989, the Nikkei 225 fell more than 60% over the next two decades, while real estate prices collapsed. Dividends in Japan were meager, leaving investors with devastating real losses. Gold priced in yen, by contrast, rose steadily, not because gold surged globally but because the yen weakened and equities collapsed. In this case, gold's role as a stabilizer of purchasing power within a failing equity market was unmistakable.
But the reverse is also true: when productive economies expand, gold falls behind. The U.S. postwar boom from 1945 to 1971, under the Bretton Woods system, left gold flat at $35 while equities multiplied many times over. Central bank policy capped gold's movement, and ordinary savers in gold missed the extraordinary returns of equities. The "different risk" profile of gold is thus evident across continents: it shines during structural crisis, but lags badly in prosperity.
Correlation evidence strengthens this point. Dirk Baur and Brian Lucey showed that gold functions as a hedge in normal times and a safe haven during systemic shocks. Their 2010 study, along with Baur and McDermott's global extension, confirmed that in crises gold's correlation with equities turns negative. Virginie Coudert and Hélène Raymond quantified this effect for U.S. recessions between 1978 and 2009: gold returned nearly 12% while equities lost 4%. In equity bear markets, gold's average loss was only 1%, compared to 20% for stocks. These findings reveal that gold is not a generator of wealth but a volatility buffer.
Yet investors cannot forget the cost of this insurance. The long decline from 1980 to 1999 shows that gold can underperform for decades. An investor who allocated heavily to gold in 1980 endured 30 years of negative or flat real returns, while equities and even bonds multiplied. This opportunity cost is immense and explains why most portfolio optimizations limit gold to 5–15% at most. Too much gold means decades of stagnation; too little leaves exposure to systemic risk.
Central banks themselves demonstrate this balance in practice. The U.S. held around 20,000 tonnes of gold in the 1950s and 1960s, but the London Gold Pool collapsed in 1968 as governments could not defend fixed prices. The Nixon shock in 1971 ended convertibility. In the 1980s and 1990s, many European banks sold gold, most famously the U.K., which sold 395 tonnes at near-bottom prices in 1999–2002. But the Washington Agreement of 1999 capped further coordinated sales, and since 2000 central banks have become net buyers.
The magnitude of these purchases underscores gold's continuing role. Russia quadrupled its reserves to over 2,300 tonnes by 2020. China quietly accumulated more than 2,000 tonnes. Turkey and India also built reserves, while Germany repatriated hundreds of tonnes from foreign vaults. According to the World Gold Council, central banks bought 1,037 tonnes in 2023, 1,086 tonnes in 2024 — a record — and nearly 600 tonnes in the first half of 2025. The Bank for International Settlements modeled optimal reserve allocations in 2020 and concluded that 10–20% gold exposure provided diversification benefits with low correlation to bonds and currencies.
For central banks, gold is not about return but about independence. It carries no counterparty risk, cannot default, and is accepted globally. Unlike Treasury bonds, which expose holders to U.S. fiscal policy, gold sits outside any single government's liability structure. This explains why, even as gold underperforms equities and often bonds, official reserves still amount to over 36,000 tonnes worldwide, roughly 17% of global reserves. It is the ultimate hedge against systemic disruption, but never a growth engine.
The lessons for private investors mirror those for central banks. Gold cannot be expected to compound like equities, or even like bonds. Its volatility is high, its Sharpe ratio near zero, and its drawdowns can last decades. But in moments of systemic crisis — inflationary spirals, fiscal collapses, geopolitical shocks — it provides unique insurance. The challenge is calibrating exposure: too much guarantees mediocrity, too little leaves vulnerability. Academic consensus across BIS, IMF, and World Gold Council research converges on the same conclusion: gold belongs in a portfolio, but always as a minority share, never the core.
Gold's relationship with deflation adds another layer of complexity. Roy Jastram, in The Golden Constant, highlighted that during deflationary episodes such as the seventeenth century or the Great Depression, gold's purchasing power actually increased. Consumer prices fell more quickly than gold's nominal value, so holders of gold could buy more goods with the same ounces. In the 1930s, for example, when U.S. consumer prices dropped by nearly 25%, gold's official peg meant its real purchasing power rose dramatically once revalued under Roosevelt. Bonds in that decade provided decent nominal returns but suffered credit stress, while equities were crushed. Gold's constancy in deflation explains why it is often considered a hedge against both inflation and deflation — though its performance differs sharply across the two.
Japan's lost decades underscore this point. From 1989 to 2009, equity and real estate values in Japan collapsed, while consumer prices stagnated or fell. Domestic investors who held cash or yen-denominated bonds preserved nominal wealth but earned little. Gold, priced in yen, appreciated steadily, reflecting both the weak domestic environment and global safe-haven demand. In this case, gold acted less as a growth asset and more as a stabilizer of purchasing power in a prolonged deflationary economy.
The paradox of gold is that it can underperform for decades even while being an excellent diversifier. Correlation studies provide the evidence. Dirk Baur and Brian Lucey demonstrated that in normal times, gold's correlation with equities hovers near zero — neither hedge nor drag. In crises, however, the correlation turns sharply negative, making gold a safe haven. Baur and McDermott extended this work globally, confirming that during systemic shocks gold decouples from equity markets everywhere. This non-linear correlation explains why gold's portfolio value exceeds what its low Sharpe ratio would otherwise justify.
But investors must pay for this insurance. Virginie Coudert and Hélène Raymond quantified that during U.S. recessions between 1978 and 2009, gold gained about 12% while equities lost 4%. In bear markets, gold fell only 1%, compared to equities' 20%. Yet over full cycles, gold underperformed equities by a wide margin, averaging 4–6% nominal versus 10–11% for stocks. Gold is thus a form of downside insurance: costly in the long run but occasionally priceless in crises.
The survivorship bias of equities highlights why gold continues to appeal. Hendrik Bessembinder's 2018 study revealed that only 4% of U.S. stocks accounted for all net wealth creation between 1926 and 2016. The other 96% either matched T-bills or destroyed wealth. For investors unlucky enough to concentrate in losers, gold's inertness would have been far superior. While equities dominate in aggregate, the journey of individual stocks can be ruinous. Gold cannot create fortunes, but it cannot go bankrupt either.
This leads naturally to the diversification studies. Colin Lawrence (2002) and later Dirk Baur and Kristoffer Glover (2023) showed that adding gold to traditional stock-bond portfolios improves risk-adjusted returns. The optimal allocation varies between 5% and 20%, depending on assumptions, but the conclusion is consistent: gold enhances portfolio resilience by lowering drawdowns. Juan Noguera and Dirk Baur (2024) quantified this, showing that portfolios with ~17% gold achieved Sharpe ratios 20% higher than stock-bond mixes. Gold itself is a weak standalone investment, but in combination, it strengthens the whole.
The behavior of central banks reinforces this conclusion. Since 2000, official institutions have become net buyers of gold, reversing the selling trend of the 1980s and 1990s. Russia, China, India, and Turkey have led this movement, while Germany repatriated hundreds of tonnes from foreign vaults. The World Gold Council's 2024 report confirmed central bank purchases of over 1,000 tonnes in both 2023 and 2024 — the largest ever. The Bank for International Settlements has modeled optimal allocations at 10–20% of reserves, reflecting gold's diversification benefit.
At the same time, gold's volatility limits its appeal. McCown and Zimmerman characterized it as a "zero-beta" asset — uncorrelated but volatile. Deutsche Bank's 2024 Long-Term Asset Return Study confirmed that gold's drawdowns are longer and deeper than equity bear markets, sometimes lasting decades. This is why institutional optimizations rarely allocate more than 20% to gold, even for central banks. Too much gold stifles growth; too little leaves vulnerability.
Comparisons with other safe assets deepen the picture. U.S. Treasury Inflation-Protected Securities (TIPS), introduced in 1997, offer direct inflation hedging with yield. Studies by Campbell and Viceira show that TIPS have outperformed gold on a risk-adjusted basis since inception, though gold remains superior in extreme geopolitical stress. Real estate, particularly farmland and timberland, has also beaten gold over the long run by combining inflation sensitivity with yield. Gold is unique in its neutrality, but not unique in protecting wealth.
Other studies highlight the poor risk-adjusted returns of gold. Barro and Misra found gold's Sharpe ratio near zero over centuries, compared to 0.3–0.4 for equities and 0.2–0.3 for bonds. UBS Yearbook data reinforce this: from 1971 to 2023, gold produced about 6–7% nominal with 15–20% volatility, compared to 11% for equities with similar volatility. On a risk-adjusted basis, equities provided far superior compensation for volatility. Gold's diversification role survives only because of its unique correlation pattern.
This pattern plays out vividly in crisis decades. In the 1970s, gold rose more than 400% while equities stagnated and bonds collapsed. In the 1980s and 1990s, gold fell relentlessly while equities soared. In the 2000s, gold rose again, climbing from $250 to $1,900 during the dot-com bust and financial crisis. In the 2010s, it stagnated again. The cycle is unmistakable: gold thrives in stress, lags in calm. Equities and bonds compound, but gold only counters volatility.
Central banks' reliance on gold proves its enduring symbolic power. The metal carries no counterparty risk, requires no issuer, and functions globally as a reserve of last resort. Unlike U.S. Treasuries, which can be frozen or defaulted upon, gold is neutral. This explains why even as fiat currencies dominate, central banks hold 36,000 tonnes of gold — roughly 17% of global reserves. In moments of fracture, gold remains the only asset outside the reach of political will.
For investors, this produces a paradox. Gold is simultaneously one of the weakest assets for long-run growth and one of the strongest for short-run protection. Its Sharpe ratio is poor, its compounding negligible, but its diversification properties justify its place in portfolios. The lesson is clear: gold is not for those seeking to multiply wealth. It is for those seeking to preserve it when everything else is breaking.
Even so, gold's track record comes with caveats. Claims of a "5,000-year track record" are more marketing than data. Archaeological evidence shows gold used as adornment in predynastic Egypt and Mesopotamia, but as true currency only from about 600 BCE with Lydian coinage. Its history is patchy and context-dependent. What has endured is not uninterrupted financial return, but cultural and symbolic value that ensures continuity of trust.
Edit:
Corrections & Clarifications
Civil War & Gold Premium
Earlier phrasing implied a simple gold "spike." Correction: During the U.S. Civil War, the greenback traded at a fluctuating premium/discount versus gold. The "Resumption Act" restored parity in 1879. Gold's purchasing power rose during post-war deflation.
Long Depression (1873–1896)
Clarified that gold's nominal price was fixed at $20.67/oz, so its purchasing power rose as prices fell 1–2% annually. Farmers and debtors suffered, while creditors and gold holders gained.
China's Monetary Standard
Correction: China remained on a silver standard, not gold, until the 1935 "fabi" reform. Silver depreciated due to global over-supply, causing domestic deflation. Gold held more stable value in trade and elite savings.
Great Depression
Correction: Roosevelt revalued gold from $20.67/oz to $35/oz in 1934 (not 1933), an overnight increase of ~69%. This revaluation, combined with ~25% CPI deflation, dramatically increased gold's purchasing power.
Japan's Lost Decades
Clarified: The Nikkei peaked at 38,915 in Dec 1989 and did not surpass that level until Feb 2024. Gold priced in yen steadily rose during Japan's deflationary stagnation, providing relative stability.
1940s Inflation & Yield Caps
Correction: During WWII, U.S. Treasury yields were pegged by the Fed (1942–1951). CPI inflation surged, eroding bonds' real value. Gold's official price remained $35/oz, though its black-market and international value rose.
1970s Stagflation
Confirmed: Gold soared from $35 in 1971 to ~$850 in Jan 1980, more than 20× nominal gains. In real terms, gold rose 400–500%, far outpacing equities and bonds.
1980–2001 Gold Slump
Earlier phrasing said "gold lost 80%." Correction: Gold fell ≈70–85% in real (inflation-adjusted) terms from 1980 ($850/oz) to 2001 (~$250/oz). In nominal terms the drop was smaller (~70%).
Suit & Bread Comparisons
Correction: "One ounce of gold = one nice suit" is a cultural rule-of-thumb, not a strict historical constant. In Rome, 1890s America, and modern times, an ounce of gold roughly covered high-quality clothing, but regional/quality variations are large. Bread prices likewise vary by place and quality — the parallels hold in broad strokes, but should not be overstated.
Bessembinder Equity Study
Clarification: Only ~4% of U.S. stocks created all net wealth 1926–2016. The rest matched or underperformed T-bills. Gold's inertness compares favorably to losing stocks, though equities dominate in aggregate.
Central Bank Purchases
Correction: WGC data show net purchases of 1,082t (2022), 1,037t (2023), and 1,086t (2024, record). Our earlier figures slightly overstated 2022.
Hyperinflation Cases
Volatility & Sharpe Ratio
Correction: Gold's long-run Sharpe ratio is near 0, with ~15–20% volatility and ~1–2% real returns. Equities achieve Sharpe ratios ~0.3–0.4 with similar volatility, making gold poor on a risk-adjusted basis.
The differences between these assets appear small when expressed as annual returns, but over decades and centuries, they diverge dramatically. Equities returning 6–7% real per year double purchasing power roughly every decade, multiplying a single dollar into more than a thousand across a century. Bonds, at 1–2% real, may turn that dollar into twenty or thirty. Bills and deposits, hovering close to zero after inflation, barely preserve capital at all. Gold, meanwhile, stands apart. Its purchasing power has shown uncanny stability, not growth. The same ounce of gold that could outfit a Roman centurion with a fine cloak and sandals, or buy a gentleman's suit in 1890s America, still buys a fine suit today. Its power is cultural continuity, not compounding.
Roy Jastram's classic The Golden Constant (1977, updated 2009 by Jill Leyland) captured this phenomenon with centuries of data from England and the United States. His results showed that gold's purchasing power mean-reverts. In inflationary times, gold lags in the short run but eventually catches up as fiat currencies devalue. In deflation, gold's purchasing power rises as consumer prices fall faster than gold's nominal value. Across the long arc of history, the real return of gold averages essentially zero. It is reliable for continuity, not growth.
More recent research confirms this. Campbell Harvey, Claude Erb, and John Welsh extended Jastram's work to 2020, analyzing gold back to 1257. They concluded that gold has returned about 0.5–1% annually in real terms, versus ~7% for U.S. equities. Robert Barro and Sanjay Misra studied gold from 1836 to 2011 and reached a similar conclusion: a 1.1% real return, dwarfed by equities but better than cash. The evidence is overwhelming: gold preserves, equities multiply.
The story of equities is one of extraordinary compounding. Jeremy Siegel's Stocks for the Long Rundocuments that U.S. equities have averaged ~6.7% real CAGR since 1802. Dimson, Marsh, and Staunton put the global figure at ~5.3% since 1900, with the U.S. achieving ~8.5%. Jorion and Goetzmann's study of 39 countries since 1921 found equities averaged ~4.3% real, but warned of survivorship bias since many markets were wiped out by wars or expropriation. The key point is that equities dominate where they endure.
The mathematics of compounding makes this dominance unavoidable. A dollar invested in U.S. equities in 1900, with dividends reinvested, would be worth more than $2,000 in real terms today. The same dollar in gold would scarcely exceed $10. Equities' superiority is structural, tied to dividends, productivity, and retained earnings. Gold may safeguard wealth, but only equities multiply it.
Dividends are central to equities' supremacy. Studies by the CFA Institute and others show that the majority of long-term equity wealth arises from reinvested dividends rather than price appreciation. Gold, in contrast, generates no cash flow. Its scarcity supports its role as a hedge, but its inertness limits its long-run performance.
This divide is the essence of the equity premium puzzle. Rajnish Mehra and Edward Prescott showed in 1985 that equities outperformed Treasury bills by ~6% annually over the twentieth century, far beyond what risk models could justify. Later work by Campbell and Shiller, and Fama and French, confirmed the persistence of this anomaly. Equities are consistently rewarded with extraordinary returns, while gold, behaving like an inert store of value, returns little beyond inflation.
Other studies reinforced this. Santa-Clara and Valkanov, Keim and Stambaugh, and the UBS Yearbooks all confirmed that equities offer a persistent, global premium. Survivorship bias explains part of this — many firms fail — but the index as a whole reflects the productivity of surviving enterprises. Gold avoids bankruptcy but offers no growth; equities are risky individually but explosive in aggregate.
Bonds occupy the middle rung. Sidney Homer and Richard Sylla's A History of Interest Ratesdocuments thousands of years of fixed-income markets, while modern data from DMS, Barclays, and Deutsche Bank show government bonds earning ~1–2% real CAGR since 1900. Robert Shiller's U.S. data back to 1871 confirm this figure. Bonds produce positive real growth but are vulnerable to inflation.
This vulnerability is clear in Antti Ilmanen's Expected Returns. Bonds thrive in disinflationary periods such as 1980–2000, when falling yields boosted prices. In inflationary decades like the 1970s, bonds were crushed in real terms, while gold soared. Over the long run, bonds provide modest compounding, while gold provides continuity.
Bills and CDs are weaker still. Data from Ibbotson and DMS show bills averaging ~0.5–0.8% real since 1900. They serve as liquid, nominally safe assets but are eroded by inflation. John Campbell and Luis Viceira argued they are useful for short-term stability but cannot preserve multigenerational wealth. Gold, despite lacking yield, performs better than cash across centuries because fiat currencies erode.
Volatility comparisons further illuminate the trade-offs. Equities and gold have both shown ~15–20% annualized volatility since 1971. Yet equities delivered ~7% real CAGR, while gold delivered close to zero. McCown and Zimmerman described gold as a "zero-beta asset" — moving independently of equities, but still volatile. Investors often misinterpret gold's safe-haven role as implying stability, but in practice gold can be just as wild as stocks.
Indeed, equities experience dramatic crashes but tend to rebound. The Great Depression erased nearly 80% of U.S. equity value, and 2008 cut markets in half. Yet dividends and earnings growth restored wealth. Gold suffers different risks: long grinding drawdowns. Between 1980 and 1999, gold lost ~70% nominally and ~80% in real terms, taking more than three decades to recover.
And gold's volatility can even surpass that of equities. Standard deviation during crisis decades has spiked above 25–30%, higher than most equity markets. Its movements are driven less by fundamentals and more by macro shocks: inflation surges, currency collapses, or changes in the monetary system. Erb and Harvey showed that gold's volatility clustered in crisis decades, with extreme swings in the 1970s, the long collapse of the 1980s–1990s, and the rebound during 2008–2011.
This profile creates an asymmetrical burden. In calm decades such as the 1980s and 1990s, equities provided rising returns while gold bled value. In crisis decades, gold's gains came with chaotic swings rivaling equities. Deutsche Bank confirmed that gold's drawdowns are often longer and its recoveries entirely dependent on rare systemic shocks. Analysts like Claude Erb stress that gold is not "low risk" but "different risk": useful for tail events, punishing in stability.
Correlation evidence helps explain why investors still allocate to gold. Dirk Baur and Brian Lucey found gold acts as a hedge on average and a safe haven in crises. Baur and McDermott extended this internationally, confirming the pattern. Virginie Coudert and Hélène Raymond quantified the effect: during U.S. recessions from 1978 to 2009, gold gained ~12% while equities lost ~4%. In bear markets, gold's losses averaged just 1%, compared to equities' 20%.
Survivorship bias in equities underscores the difference. Hendrik Bessembinder found in 2018 that just 4% of stocks accounted for all net wealth creation in the U.S. from 1926 to 2016. The vast majority matched T-bills or destroyed wealth. Gold, though unproductive, cannot fail outright. It lacks upside but avoids the risk of total ruin that haunts individual equities.
One of the clearest ways to illustrate gold's weakness as a long-term investment is through its Sharpe ratio, the standard measure of risk-adjusted returns. Since the collapse of Bretton Woods in 1971, gold has returned about 6–7% nominal per year but with annualized volatility of roughly 15–20%. Adjusted for inflation, its real return is closer to 1–2%, producing a Sharpe ratio near zero over long horizons. By contrast, U.S. equities in the same period have delivered real returns of 6–7% with similar volatility, giving them a Sharpe ratio closer to 0.3–0.4, depending on the sample window (Dimson, Marsh, Staunton; UBS Yearbook 2023). Bonds, though lower-returning, often achieve Sharpe ratios in the 0.2–0.3 range because of reduced volatility. In other words, every unit of risk taken in gold has historically produced little or no incremental reward, confirming Claude Erb and Campbell Harvey's conclusion that gold's primary utility lies not in its risk-adjusted performance but in its diversification properties during crises.
Gold's drawdown history is one of the strongest reminders that volatility does not mean guaranteed growth. After peaking in 1980 near $850/oz, gold spent almost two decades in decline. By 1999 it had fallen to about $250/oz, representing a 70% nominal loss and closer to 80% in real terms once inflation was considered. This slump lasted longer than any single equity bear market in modern history, and only by 2011 did gold finally reclaim and surpass its 1980 highs. Equities, by contrast, while suffering devastating collapses in the Great Depression (nearly –80%) and again in 2008 (–50%), recovered through earnings growth and dividends. Gold has no such recovery mechanism; it must wait for the next crisis or inflationary surge to find support.
This contrast is central to the psychology of investors. Equities are tied to the productive capacity of firms. As long as profits return, equity markets recover, and those who reinvest dividends are rewarded. Gold, however, offers no organic growth. Its long periods of stagnation, like the 1980–1999 slump, create opportunity costs that can devastate long-term savers if it is treated as a primary investment. This is why scholars like Claude Erb and Campbell Harvey describe gold not as a growth asset but as "different risk," one that hedges specific scenarios but often fails in normal conditions.
Yet bonds provide a counterpoint, offering moderate but genuine compounding. Long-term government bonds have delivered around 2% real annual returns since 1900, according to the Dimson–Marsh–Staunton database and Robert Shiller's U.S. series. But bonds' strength depends heavily on the regime. In the 1970s, when inflation averaged over 7% per year, bondholders suffered losses exceeding 40% in real terms, while gold rose more than 400%. In the disinflationary 1980–2000 era, however, bonds produced their best century-long run, as yields fell from double digits to below 5%, while gold collapsed. Antti Ilmanen's Expected Returns illustrates this "mirror image": gold thrives when inflation erodes fixed-income assets, but bonds thrive when inflation is tamed.
Historical episodes show this dynamic clearly. During the U.S. Civil War, the government issued "greenbacks," paper notes not fully backed by gold. Bonds tied to these notes plummeted in value as inflation surged, while gold became the black-market medium of choice for merchants. Homer and Sylla's History of Interest Rates notes that gold premiums in New York reached nearly 200% at the war's peak. In both World Wars, government bonds were subject to yield caps and inflation, meaning their real returns were negative. Gold, though formally pegged, was hoarded in private markets and held its value more effectively. These episodes confirm that gold serves best as a hedge against extreme fiscal stress, while bonds are reliable in stability.
Bills and short-term deposits occupy a still weaker position. Ibbotson's SBBI Yearbook and DMS data show Treasury bills averaging only 0.5–0.8% real returns since 1900. In calm times they serve their purpose — providing liquidity, stability, and a nominally safe asset. But in inflationary or wartime settings, they are destroyed in real terms. During the Great Depression, Treasury bills preserved nominal value but produced near-zero real return. Gold, though fixed in U.S. law at $20.67/oz, was revalued by Franklin Roosevelt in 1933 to $35, a 69% jump that handed holders a windfall while bills sat stagnant.
The World War II era illustrates this divergence again. Treasury yields were held artificially low by government policy, often below 2%. Inflation, however, ran higher, wiping out real gains. Gold, though officially fixed, retained underground value and provided continuity of wealth across borders. In practice, bills eroded while gold endured. Modern certificates of deposit tell a similar story. In the early 1980s, nominal CD rates touched 15%, but inflation was nearly as high, leaving real returns modest. Over the long run, cash-like instruments lag every other asset class, beaten even by inert gold.
The volatility comparison sharpens when extended globally. Japan's "lost decades" offer an instructive case. After peaking in 1989, the Nikkei 225 fell more than 60% over the next two decades, while real estate prices collapsed. Dividends in Japan were meager, leaving investors with devastating real losses. Gold priced in yen, by contrast, rose steadily, not because gold surged globally but because the yen weakened and equities collapsed. In this case, gold's role as a stabilizer of purchasing power within a failing equity market was unmistakable.
But the reverse is also true: when productive economies expand, gold falls behind. The U.S. postwar boom from 1945 to 1971, under the Bretton Woods system, left gold flat at $35 while equities multiplied many times over. Central bank policy capped gold's movement, and ordinary savers in gold missed the extraordinary returns of equities. The "different risk" profile of gold is thus evident across continents: it shines during structural crisis, but lags badly in prosperity.
Correlation evidence strengthens this point. Dirk Baur and Brian Lucey showed that gold functions as a hedge in normal times and a safe haven during systemic shocks. Their 2010 study, along with Baur and McDermott's global extension, confirmed that in crises gold's correlation with equities turns negative. Virginie Coudert and Hélène Raymond quantified this effect for U.S. recessions between 1978 and 2009: gold returned nearly 12% while equities lost 4%. In equity bear markets, gold's average loss was only 1%, compared to 20% for stocks. These findings reveal that gold is not a generator of wealth but a volatility buffer.
Yet investors cannot forget the cost of this insurance. The long decline from 1980 to 1999 shows that gold can underperform for decades. An investor who allocated heavily to gold in 1980 endured 30 years of negative or flat real returns, while equities and even bonds multiplied. This opportunity cost is immense and explains why most portfolio optimizations limit gold to 5–15% at most. Too much gold means decades of stagnation; too little leaves exposure to systemic risk.
Central banks themselves demonstrate this balance in practice. The U.S. held around 20,000 tonnes of gold in the 1950s and 1960s, but the London Gold Pool collapsed in 1968 as governments could not defend fixed prices. The Nixon shock in 1971 ended convertibility. In the 1980s and 1990s, many European banks sold gold, most famously the U.K., which sold 395 tonnes at near-bottom prices in 1999–2002. But the Washington Agreement of 1999 capped further coordinated sales, and since 2000 central banks have become net buyers.
The magnitude of these purchases underscores gold's continuing role. Russia quadrupled its reserves to over 2,300 tonnes by 2020. China quietly accumulated more than 2,000 tonnes. Turkey and India also built reserves, while Germany repatriated hundreds of tonnes from foreign vaults. According to the World Gold Council, central banks bought 1,037 tonnes in 2023, 1,086 tonnes in 2024 — a record — and nearly 600 tonnes in the first half of 2025. The Bank for International Settlements modeled optimal reserve allocations in 2020 and concluded that 10–20% gold exposure provided diversification benefits with low correlation to bonds and currencies.
For central banks, gold is not about return but about independence. It carries no counterparty risk, cannot default, and is accepted globally. Unlike Treasury bonds, which expose holders to U.S. fiscal policy, gold sits outside any single government's liability structure. This explains why, even as gold underperforms equities and often bonds, official reserves still amount to over 36,000 tonnes worldwide, roughly 17% of global reserves. It is the ultimate hedge against systemic disruption, but never a growth engine.
The lessons for private investors mirror those for central banks. Gold cannot be expected to compound like equities, or even like bonds. Its volatility is high, its Sharpe ratio near zero, and its drawdowns can last decades. But in moments of systemic crisis — inflationary spirals, fiscal collapses, geopolitical shocks — it provides unique insurance. The challenge is calibrating exposure: too much guarantees mediocrity, too little leaves vulnerability. Academic consensus across BIS, IMF, and World Gold Council research converges on the same conclusion: gold belongs in a portfolio, but always as a minority share, never the core.
Gold's relationship with deflation adds another layer of complexity. Roy Jastram, in The Golden Constant, highlighted that during deflationary episodes such as the seventeenth century or the Great Depression, gold's purchasing power actually increased. Consumer prices fell more quickly than gold's nominal value, so holders of gold could buy more goods with the same ounces. In the 1930s, for example, when U.S. consumer prices dropped by nearly 25%, gold's official peg meant its real purchasing power rose dramatically once revalued under Roosevelt. Bonds in that decade provided decent nominal returns but suffered credit stress, while equities were crushed. Gold's constancy in deflation explains why it is often considered a hedge against both inflation and deflation — though its performance differs sharply across the two.
Japan's lost decades underscore this point. From 1989 to 2009, equity and real estate values in Japan collapsed, while consumer prices stagnated or fell. Domestic investors who held cash or yen-denominated bonds preserved nominal wealth but earned little. Gold, priced in yen, appreciated steadily, reflecting both the weak domestic environment and global safe-haven demand. In this case, gold acted less as a growth asset and more as a stabilizer of purchasing power in a prolonged deflationary economy.
The paradox of gold is that it can underperform for decades even while being an excellent diversifier. Correlation studies provide the evidence. Dirk Baur and Brian Lucey demonstrated that in normal times, gold's correlation with equities hovers near zero — neither hedge nor drag. In crises, however, the correlation turns sharply negative, making gold a safe haven. Baur and McDermott extended this work globally, confirming that during systemic shocks gold decouples from equity markets everywhere. This non-linear correlation explains why gold's portfolio value exceeds what its low Sharpe ratio would otherwise justify.
But investors must pay for this insurance. Virginie Coudert and Hélène Raymond quantified that during U.S. recessions between 1978 and 2009, gold gained about 12% while equities lost 4%. In bear markets, gold fell only 1%, compared to equities' 20%. Yet over full cycles, gold underperformed equities by a wide margin, averaging 4–6% nominal versus 10–11% for stocks. Gold is thus a form of downside insurance: costly in the long run but occasionally priceless in crises.
The survivorship bias of equities highlights why gold continues to appeal. Hendrik Bessembinder's 2018 study revealed that only 4% of U.S. stocks accounted for all net wealth creation between 1926 and 2016. The other 96% either matched T-bills or destroyed wealth. For investors unlucky enough to concentrate in losers, gold's inertness would have been far superior. While equities dominate in aggregate, the journey of individual stocks can be ruinous. Gold cannot create fortunes, but it cannot go bankrupt either.
This leads naturally to the diversification studies. Colin Lawrence (2002) and later Dirk Baur and Kristoffer Glover (2023) showed that adding gold to traditional stock-bond portfolios improves risk-adjusted returns. The optimal allocation varies between 5% and 20%, depending on assumptions, but the conclusion is consistent: gold enhances portfolio resilience by lowering drawdowns. Juan Noguera and Dirk Baur (2024) quantified this, showing that portfolios with ~17% gold achieved Sharpe ratios 20% higher than stock-bond mixes. Gold itself is a weak standalone investment, but in combination, it strengthens the whole.
The behavior of central banks reinforces this conclusion. Since 2000, official institutions have become net buyers of gold, reversing the selling trend of the 1980s and 1990s. Russia, China, India, and Turkey have led this movement, while Germany repatriated hundreds of tonnes from foreign vaults. The World Gold Council's 2024 report confirmed central bank purchases of over 1,000 tonnes in both 2023 and 2024 — the largest ever. The Bank for International Settlements has modeled optimal allocations at 10–20% of reserves, reflecting gold's diversification benefit.
At the same time, gold's volatility limits its appeal. McCown and Zimmerman characterized it as a "zero-beta" asset — uncorrelated but volatile. Deutsche Bank's 2024 Long-Term Asset Return Study confirmed that gold's drawdowns are longer and deeper than equity bear markets, sometimes lasting decades. This is why institutional optimizations rarely allocate more than 20% to gold, even for central banks. Too much gold stifles growth; too little leaves vulnerability.
Comparisons with other safe assets deepen the picture. U.S. Treasury Inflation-Protected Securities (TIPS), introduced in 1997, offer direct inflation hedging with yield. Studies by Campbell and Viceira show that TIPS have outperformed gold on a risk-adjusted basis since inception, though gold remains superior in extreme geopolitical stress. Real estate, particularly farmland and timberland, has also beaten gold over the long run by combining inflation sensitivity with yield. Gold is unique in its neutrality, but not unique in protecting wealth.
Other studies highlight the poor risk-adjusted returns of gold. Barro and Misra found gold's Sharpe ratio near zero over centuries, compared to 0.3–0.4 for equities and 0.2–0.3 for bonds. UBS Yearbook data reinforce this: from 1971 to 2023, gold produced about 6–7% nominal with 15–20% volatility, compared to 11% for equities with similar volatility. On a risk-adjusted basis, equities provided far superior compensation for volatility. Gold's diversification role survives only because of its unique correlation pattern.
This pattern plays out vividly in crisis decades. In the 1970s, gold rose more than 400% while equities stagnated and bonds collapsed. In the 1980s and 1990s, gold fell relentlessly while equities soared. In the 2000s, gold rose again, climbing from $250 to $1,900 during the dot-com bust and financial crisis. In the 2010s, it stagnated again. The cycle is unmistakable: gold thrives in stress, lags in calm. Equities and bonds compound, but gold only counters volatility.
Central banks' reliance on gold proves its enduring symbolic power. The metal carries no counterparty risk, requires no issuer, and functions globally as a reserve of last resort. Unlike U.S. Treasuries, which can be frozen or defaulted upon, gold is neutral. This explains why even as fiat currencies dominate, central banks hold 36,000 tonnes of gold — roughly 17% of global reserves. In moments of fracture, gold remains the only asset outside the reach of political will.
For investors, this produces a paradox. Gold is simultaneously one of the weakest assets for long-run growth and one of the strongest for short-run protection. Its Sharpe ratio is poor, its compounding negligible, but its diversification properties justify its place in portfolios. The lesson is clear: gold is not for those seeking to multiply wealth. It is for those seeking to preserve it when everything else is breaking.
Even so, gold's track record comes with caveats. Claims of a "5,000-year track record" are more marketing than data. Archaeological evidence shows gold used as adornment in predynastic Egypt and Mesopotamia, but as true currency only from about 600 BCE with Lydian coinage. Its history is patchy and context-dependent. What has endured is not uninterrupted financial return, but cultural and symbolic value that ensures continuity of trust.
Edit:
Corrections & Clarifications
Civil War & Gold Premium
Earlier phrasing implied a simple gold "spike." Correction: During the U.S. Civil War, the greenback traded at a fluctuating premium/discount versus gold. The "Resumption Act" restored parity in 1879. Gold's purchasing power rose during post-war deflation.
Long Depression (1873–1896)
Clarified that gold's nominal price was fixed at $20.67/oz, so its purchasing power rose as prices fell 1–2% annually. Farmers and debtors suffered, while creditors and gold holders gained.
China's Monetary Standard
Correction: China remained on a silver standard, not gold, until the 1935 "fabi" reform. Silver depreciated due to global over-supply, causing domestic deflation. Gold held more stable value in trade and elite savings.
Great Depression
Correction: Roosevelt revalued gold from $20.67/oz to $35/oz in 1934 (not 1933), an overnight increase of ~69%. This revaluation, combined with ~25% CPI deflation, dramatically increased gold's purchasing power.
Japan's Lost Decades
Clarified: The Nikkei peaked at 38,915 in Dec 1989 and did not surpass that level until Feb 2024. Gold priced in yen steadily rose during Japan's deflationary stagnation, providing relative stability.
1940s Inflation & Yield Caps
Correction: During WWII, U.S. Treasury yields were pegged by the Fed (1942–1951). CPI inflation surged, eroding bonds' real value. Gold's official price remained $35/oz, though its black-market and international value rose.
1970s Stagflation
Confirmed: Gold soared from $35 in 1971 to ~$850 in Jan 1980, more than 20× nominal gains. In real terms, gold rose 400–500%, far outpacing equities and bonds.
1980–2001 Gold Slump
Earlier phrasing said "gold lost 80%." Correction: Gold fell ≈70–85% in real (inflation-adjusted) terms from 1980 ($850/oz) to 2001 (~$250/oz). In nominal terms the drop was smaller (~70%).
Suit & Bread Comparisons
Correction: "One ounce of gold = one nice suit" is a cultural rule-of-thumb, not a strict historical constant. In Rome, 1890s America, and modern times, an ounce of gold roughly covered high-quality clothing, but regional/quality variations are large. Bread prices likewise vary by place and quality — the parallels hold in broad strokes, but should not be overstated.
Bessembinder Equity Study
Clarification: Only ~4% of U.S. stocks created all net wealth 1926–2016. The rest matched or underperformed T-bills. Gold's inertness compares favorably to losing stocks, though equities dominate in aggregate.
Central Bank Purchases
Correction: WGC data show net purchases of 1,082t (2022), 1,037t (2023), and 1,086t (2024, record). Our earlier figures slightly overstated 2022.
Hyperinflation Cases
- Weimar Germany (1921–23): Gold preserved purchasing power as the mark collapsed (170 marks/oz → 87 trillion marks/oz).
- Russia (1920s): Hyperinflation destroyed the ruble; the gold-backed chervonets (1922) restored confidence.
- Zimbabwe (2000s): Inflation at 89.7 sextillion %; gold dust was used in markets.
- Venezuela (2010s–20s): Gold jewelry/jewels preserved value amid bolívar collapse.
- Turkey (2018–25): Lira lost ~80% in 5 years; gold widely adopted in households.
- Lebanon (2019–25): Pound lost 95%; gold held value when banks froze accounts.
- Vietnam (1980s–90s): Inflation >400%; gold became the preferred medium for real estate and large transactions.
Volatility & Sharpe Ratio
Correction: Gold's long-run Sharpe ratio is near 0, with ~15–20% volatility and ~1–2% real returns. Equities achieve Sharpe ratios ~0.3–0.4 with similar volatility, making gold poor on a risk-adjusted basis.
Last edited:
