Does Bitcoin Have Value? — Digital Gold, or a Giant Bubble

The debate surrounding Bitcoin begins with two conflicting assertions. Bitcoin is a digital currency sent and received on the internet without a bank. Transactions are recorded on a public ledger maintained together by participants. The principles of that ledger are covered in "What Is a Blockchain — Why a Ledger With No One in Charge Holds Up".

One side calls Bitcoin "digital gold." They say it is scarce because the total issuance is capped at 21 million. Since central banks cannot print it at will, it is a new safe-haven asset. The argument is that it protects assets from inflation (a phenomenon where prices rise overall) and government control.

The other side calls it a "giant bubble." A bubble is a phenomenon where prices rise beyond actual value and then burst. This side views it as crowd expectations piled onto numbers with no intrinsic value (the value an asset generates on its own). That is why they say the price skyrockets and then periodically collapses. In the end, they argue, it is merely a greater fool game passing the bag to someone else.

This article traces the path between those two assertions through research papers. First, there is something to clarify. Almost all papers cited here are preprints or working papers. A preprint (arXiv) is a paper made public before undergoing peer review by fellow researchers, and a working paper is an academic draft before formal presentation. Because data periods, methods, and samples vary, the conclusion of a single paper should not be read as the definitive answer as is.

The Value Debate — Does It Have Intrinsic Value?

Traditional assets have an anchor of value. Stocks have dividends and earnings, bonds have interest, and real estate has rent. This anchor is precisely intrinsic value. What does Bitcoin have?

The most sober diagnosis is the side that says "nothing." Mining is the act of spending computer calculations to record transactions on a ledger and receiving coins as a reward. The hashrate is the total amount of computational power miners put in.

One economic model matched consumers' demand for transactions with miners' supply of hashrate. This simple model assumes consumers buy only for transactions and do not hoard. In this model, prices are not determined by market equilibrium. The authors saw this as an argument. It supports the hypothesis that Bitcoin prices have no economic fundamentals and fluctuate according to holding and speculative demand (arXiv:2211.07035). Fundamentals are the basic economic factors that explain prices. Some studies also call the supply and demand of coins by this name, alongside anchors of value like dividends and interest.

There are similar theoretical analyses. Asset flow equations and experimental market analyses view typical cryptocurrencies as having no tangible value. The prediction is that in theory, prices will converge to liquidity value. Liquidity value is the total amount of cash available for purchases divided by the number of coin units. The authors' judgment is that without a link connecting to value, it is difficult to stabilize in its current form (arXiv:1802.09959).

However, saying "it has no value at all" is different from saying "there are no factors making the price." Measurable forces that move prices have been confirmed in opposing studies.

• Transaction demand is the demand to pay for goods, and speculative demand is the demand to buy seeing that the price will rise. In GARCH analysis, both had a statistically significant impact on price formation (arXiv:1812.09452).
• Even in daily data from 2009 to 2014, supply and demand fundamentals and the appeal to investors had a significant impact on prices. However, the result of prior research that macroeconomic financial indicators drive Bitcoin prices was not supported (arXiv:1405.4498).
• The most recent (2026) analysis revealed a "dual nature" where both fundamentals and speculative motives move prices. This is the result of modeling both on-chain (on the blockchain ledger) and off-chain (off the ledger) supply and demand factors together. About three quarters of transactions take place off-chain. Off-chain demand pressure had a significant impact on prices in the long run, and on-chain demand pressure in both the short and long term (arXiv:2602.08429).

An old conventional wisdom viewing mining costs as a price floor was also put to the test. The conventional wisdom is that "prices cannot drop below the mining cost." There is research that the marginal cost of production (the cost of making one more coin) is an important factor explaining prices. The valuation model is evaluated as robust (arXiv:1805.07610). Econometric analyses previously showed that mining costs do not lead prices but follow them. Other studies explain the reason with economic theory (arXiv:2204.13102). According to this explanation, costs are closer to a shadow following the price rather than a price floor.

There are also attempts to dissect the identity of value differently. There is a theory comparing the value of cryptocurrencies that can be replicated (forked) to the art market. The proposal is to divide value into "artistic value" and "use value (utility)" (arXiv:1904.05028). There is also a conclusion that Bitcoin's value comes from scarcity, utility, and public trust. The author cites the "greater fool theory" and the decreasing supply available for mining as reasons. Therefore, they forecast that Bitcoin can transition from short-term speculative investment to a more stable industry (arXiv:2010.01337). Models showing that prices move alongside people's psychological and confidence indicators have also been repeatedly presented (arXiv:1702.00215, arXiv:1709.08621).

To sum it up: There is generally no disagreement that traditional anchors of value like dividends and interest are absent. However, counterarguments that "it has intrinsic value" remain, such as research viewing marginal cost of production as a fundamental (arXiv:1805.07610). Apart from that, non-traditional price drivers such as scarcity, networks, on-chain demand, and confidence are measured. The two camps are touching different parts of the same elephant.

The Analogy of Digital Gold

The phrase "Bitcoin is digital gold" is not simple marketing. Some studies provide grounds for that analogy. A store of value is an asset that retains its value over time, and gold is the representative example of this.

One preprint backtested Bitcoin and Litecoin using on-chain metrics. The results supported the Litecoin creator's analogy of "Bitcoin equals gold, Litecoin equals silver." Bitcoin emerged as a superior store of value to Litecoin, and comparisons with gold are absent from this study (arXiv:2308.00013). There is also a suggestion that Bitcoin could one day become "sound money." At the same time, it points out two flaws in current Bitcoin. High energy consumption and cryptography vulnerable to quantum attacks. Thus, the proposal is that a redesign is necessary (arXiv:2212.13986).

However, there is research that evaluated other functions oppositely. This is a study comparing the volatility of Bitcoin, gold, and the S&P 500 (a major U.S. stock index). Volatility is the range by which prices swing. Bitcoin was a unique asset possessing both high volatility and high predictability at the same time. That volatility is primarily an intraday and intraweek phenomenon, so it softens significantly on a weekly basis. Therefore, this study viewed it as a potentially attractive store of value for investors and speculators. However, it concluded that due to its high standard deviation, it is unsuitable for use as a currency (arXiv:2103.00395). The two studies agree on it as a store of value. What diverges is the next step: "Can it be used as a currency?"

Bubbles Repeat — Price Trajectories That Rise and Burst

The core of Bitcoin skepticism is the price trajectory itself. Explosive rises followed by sudden collapses repeat periodically. This part has a relatively clear consensus. Bitcoin has repeatedly created bubbles.

The most sophisticated tool for measuring bubbles is the Log-Periodic Power Law Singularity (LPPLS) model. It captures patterns where prices accelerate unsustainably due to feedback loops like herding and imitation. In Bitcoin prices from 2012 to 2018, 3 large bubbles and 10 smaller bubbles are identified. In the large bubbles, "super-exponential" growth is confirmed. This is unsustainable growth where the rate of acceleration itself increases as prices go up (arXiv:1804.06261, arXiv:1803.05663).

Metcalfe's law states that the value of a network is proportional to the square of the number of users. Bubbles that vastly exceeded the fundamental value calculated by this law inflated and burst at least four times. LPPLS could warn of collapse risks in advance as probability intervals. However, the authors themselves drew a line, noting that the exact timing of the collapse and the trigger cannot be predicted. This is because they are factors outside the model (exogenous) (arXiv:1803.05663). With daily data, the LPPLS confidence indicator (the proportion where bubble signs were detected) failed to give warnings during short-term volatile phases. Therefore, a method was proposed to scale down to 1-hour and 30-minute intervals to detect even short-term bubbles. The authors reported that this method was effective in detecting bubbles and predicting collapses in daily data from the past two years (arXiv:1905.09647).

One study applied and refined the "social bubble hypothesis" to Bitcoin. This hypothesis views bubbles as an essential element of the technological innovation process. The authors argue that the hierarchy of repeatedly growing, exponentially larger bubbles and hype cycles bootstrapped into existence Bitcoin (arXiv:2005.06464). There are further analyses dealing with bubbles.

• In a multi-agent model, adaptive investment strategies in a infinitely divisible commodity are said to produce self-sustaining bubbles (arXiv:2012.14860).
• The trading network from December 2011 to December 2013 was analyzed. The authors found that hubs played a key role in the collapse of the second of the three bubbles during this period. Hubs are the most connected nodes, and they are said to have increased systemic risk during the bubble phase (arXiv:1805.04460).

Some studies expand beyond Bitcoin to cryptocurrencies as a whole. The correlation between the prices of native cryptocurrencies issued by metaverse projects and virtual real estate (LAND) prices exceeded 0.96. Cryptocurrency prices sometimes moved ahead of LAND prices. The study viewed this as evidence of a wealth effect, meaning that wealth expanded by cryptocurrencies increased demand for other assets. This pattern was evaluated as reminiscent of the 1920s US real estate bubble that preceded the 1929 stock market crash (arXiv:2209.04385).

There are also cases where prices did not go to zero. In a marginal cost model study, a bubble that began in the fall of 2017 was resolved as it converged to the marginal cost model in early 2018. The author concluded that even when bubbles appear, prices tend to regress to a lower bound rather than collapsing to zero (arXiv:1805.07610).

Extreme Volatility

Aside from bubbles, Bitcoin's volatility itself is a research topic. One study compares three types of volatility: historical volatility (standard deviation of past prices), GARCH-forecasted volatility, and implied volatility from the options market. Measured in all three ways, daily and annual volatility are noticeably high. However, the authors add a caveat that because liquidity in the Bitcoin options market is shallow, the implied volatility values may be distorted (arXiv:2401.02049). Realized volatility is a value calculated from actually observed price movements. From the perspective of realized volatility, there is also a diagnosis that the Bitcoin market is "an extremely risky market intertwined with volatility, jumps, and consecutive jumps" (arXiv:1912.05228).

Multiple studies have confirmed that the log values of Bitcoin volatility are "rough" (arXiv:1904.12346, arXiv:2511.03314). Being rough means that the Hurst exponent (an index where a smaller value indicates that volatility movements are more erratic) is less than 1/2. In another study examining realized volatility created from minute-by-minute prices, all finite estimates were also less than 1/2. However, the magnitude of the estimates and whether they could be calculated varied depending on the period and measurement design. Some fixed-window samples also showed multifractal traces (arXiv:2507.00575). There is also an analysis that as the market becomes more efficient, the asymmetry of volatility weakens (arXiv:2102.07425).

The impact of regulated Bitcoin futures on volatility varies across studies. One study states that CME products more than doubled volatility even after controlling for trading volume effects (arXiv:2109.15052). Conversely, another study reports that volatility grew right after introduction but became rather more stable over time (arXiv:1906.03430).

Is the Market Efficient — If Prices Already Contain All Information

Does the phrase "bubbles repeat" mean that the market is inefficient? The efficient market hypothesis (EMH) is the hypothesis that all information is already reflected in prices, making excess returns impossible. Excess returns are returns that exceed the market average. The conclusion from this perspective is nuanced.

• The Bitcoin market generally moves toward efficiency. Under specific conditions with information asymmetry or behavioral anomalies, arbitrage opportunities occasionally arise. However, it is difficult to systematically capture them. This is the conclusion of research verifying predictability using machine learning and time series (arXiv:2504.18982).
• The adaptive market hypothesis (AMH) views market efficiency as changing depending on the environment. Viewed through this hypothesis, the level of efficiency has changed over time. Bitcoin was generally more efficient than Ethereum, and efficiency has evolved over time in markets with high liquidity. The authors concluded that this result supports the adaptive market hypothesis in the well-established cryptocurrency market (arXiv:1904.09403).
• The efficiency of over 400 cryptocurrencies was measured across time windows. A market was considered efficient if patterns of price change were indistinguishable from randomly shuffled data. As a result, 37% were efficient for over 80% of the period. On the other hand, 20% were efficient for less than 20% of the period. This study summarizes that the market is heading toward "adulthood" (arXiv:1901.04967).

Meanwhile, a preprint tested whether excess returns can be generated by exploiting inefficiencies. It was tested using daily data from November 2015 to April 2018 for 1,681 cryptocurrencies. It is reported that a simple trading strategy coupled with machine learning outperformed the standard benchmark. This is a result tested on past data, and the authors go only as far as saying that it helps predict short-term changes (arXiv:1805.08550).

The Two Promises of "Digital Gold" — Inflation Hedge and Safe Haven

When Bitcoin is called "digital gold", the core promises are two. It protects assets from inflation (inflation hedge), and it becomes a shelter that holds value during crises (safe haven). A hedge is an action that reduces risk by offsetting losses in other assets.

Let's look at the inflation hedge first. A study with a blunt title concluded that "Bitcoin does not hedge inflation" (arXiv:2301.10117). Looking at the response to macroeconomic data announcements, Bitcoin reacted negatively to inflation surprises. A 1-standard-deviation surprise caused the price to drop by 24 bp (0.24 percentage points). This contradicts the practitioner consensus that Bitcoin is an inflation hedge.

There is an even stronger counterargument. A 2021 commentary from a quantitative finance perspective argued that Bitcoin in its version at the time was neither a currency nor a short- or long-term store of value. The latter refers to the assessment that the expected value does not exceed zero. It stated that it is not a reliable inflation hedge either. Furthermore, it concluded that it is neither a safe haven for investors, a shield against government tyranny, nor a tail risk defense tool against catastrophic events (arXiv:2106.14204).

The safe haven claim is a bit more complex. Studies testing whether it becomes a safe haven during crises generally viewed Bitcoin as a "weak" safe haven. This means it does not drop alongside stocks during crises, but neither does it make up for the losses.

• As Bitcoin gained attention as a safe haven, a preprint tested its safe haven and hedging properties against the US stock market index. The safe haven properties varied by period and were mainly weak safe havens in both the short and long term. In the same study, gold and silver lost their safe haven properties over time (arXiv:1703.00308).
• A negative correlation appeared between Bitcoin and Ether and stocks, bonds, and crude oil. This is a tendency where when one side goes up, the other goes down. Thus, it suggests that it can serve as a hedging tool, and putting it in a portfolio improved efficiency. However, based on tail independence—which measures the degree to which they move together during extreme downturns—it was weak as a safe haven (arXiv:1707.07977).

In other words, the strong claim that "it is as reliable as gold during crises" is not well supported by the data. At best, it is close to weak and conditional protection.

Stablecoins — Meeting New Risks While Trying to Avoid Volatility

Attempts to bypass Bitcoin's extreme volatility led to stablecoins. These are coins that claim to peg their value 1:1 to the US dollar and other assets. However, some designs have brought in a new kind of risk. The process by which algorithmic stablecoins collapsed is covered in "The Code Did Exactly What It Promised — What Algorithmic Stablecoins Guaranteed, and What They Could Not".

This Systematization of Knowledge (SoK, a paper that organizes existing research) analyzed 157 studies, 95 coins, and 44 security incidents. The summary is this. The authors viewed that stability is best understood not as an intrinsic property of a coin, but as a fragile, emergent state that relies on market trust and liquidity. Designs specialize and shift risk to specific domains rather than eliminating it. Yield mechanisms create a "dual mandate" that conflicts with stability. However, they also noted that designs reflecting lessons from past incidents had clearly lower evaluated risks (arXiv:2506.17622).

• Classifying 24 projects, over 91% of projects claimed a 1:1 peg. Meanwhile, the authors pointed out that 38% of the total use design combinations that can become vulnerable to speculative economic attacks (arXiv:1905.11905).
• Price window design is a method of setting the prices for minting and redeeming coins at fixed levels. This theoretical analysis shows that without a secondary stabilization mechanism, this approach cannot achieve both short-term and long-term stability together. It carries the risk that repeated arbitrage by speculators will eventually exhaust reserve assets. Alternatively, by widening the gap between the two prices to prevent arbitrage, the coin instead inherits fluctuations that can be proportional to the volatility of the reserve assets. This is an exception if the reserve assets themselves are already stable (arXiv:2602.15981).

It was also confirmed that risks differ depending on the design type. The tail risk spillover of 8 major stablecoins was analyzed. Fiat-collateralized types acted as an "anchor of stability." In contrast, algorithmic and crypto-collateral-backed types became "risk amplifiers" in extreme situations. Regulatory capital buffers were also addressed by the authors. Non-fiat-backed stablecoins are evaluated as needing buffers against extreme losses that must be 2 to 3 times higher than what median-based measurements suggest (arXiv:2602.18820).

There are also two related analyses. One is a copula analysis. In this analysis, stablecoin trading volume and upward volatility within the sample were linked to the volatility of the broader cryptocurrency market. The authors interpreted this as meaning stablecoins act as "dry powder" (arXiv:2603.23480). The other is a model analysis using the 2023 SVB collapse and USDC's depegging as a stress scenario. In this model, a central bank anchor reduced liquidity vulnerability (arXiv:2510.10469).

An Honest Current Status — The Evidence Conflicts

So, is Bitcoin digital gold or a bubble? The honest answer is "the evidence is mixed, and we cannot say for sure."

On the value side, the diagnostic backbone of skepticism is that there is no traditional anchor for value (arXiv:2211.07035, arXiv:1802.09959).

Advocacy points to measurement evidence showing that non-traditional drivers actually create prices (arXiv:2602.08429, arXiv:1812.09452, arXiv:1405.4498).

On the bubble side, there is relatively solid agreement on repeated bubbles and crashes (arXiv:1804.06261, arXiv:1803.05663, arXiv:2005.06464).

Yet there is a counterargument that, rather than ending at zero every time, bubbles have sometimes retreated to the marginal cost line of mining. Along with this is the rebuttal that the cost line does not prop up the price, but merely follows it (arXiv:1805.07610 vs arXiv:2204.13102).

In the cited studies, claims of an inflation hedge found no support. The basis for this is one preprint and one commentary that looked at reactions to inflation announcements. Claims of a safe haven were supported mostly at a "weak" level (arXiv:2301.10117, arXiv:1703.00308), while a commentary argued they were not a safe haven at all (arXiv:2106.14204). The two promises of digital gold sit on the weaker side within this evidence.

Ultimately, in the cited studies, Bitcoin's price actually fell in response to inflation surprises, and during crises it amounted to a weak refuge. However, it showed a negative correlation with stocks, bonds, and crude oil, suggesting it can serve as a hedging tool (arXiv:2301.10117, arXiv:1707.07977). Among stablecoins, algorithmic and crypto-collateralized types were shown to amplify risk in extreme situations (arXiv:2506.17622, arXiv:2602.18820).

The reason the conclusion is not definitive lies in the state of the data. For the exact same asset, whether it has "fundamentals/no fundamentals," is "efficient/inefficient," or is a "safe haven/not a safe haven" splits depending on the sample period and methodology. The claim of "digital gold" becomes true or false depending on the timing and conditions. The fact that the stage for all these tests was largely preprints prior to peer review remains a warning that we cannot treat the conclusion of any single paper as the definitive answer.

This article is for informational purposes only and does not constitute investment advice or solicitation. Many of the cited sources are preprints (arXiv) or working papers prior to peer review, and details may be updated by subsequent research.

Source Papers (arXiv)

Value and Intrinsic Value Debate

• On-chain and off-chain supply-demand drivers move prices (dual nature) (arXiv:2602.08429)
• Bitcoin has no economic fundamentals and fluctuates based on speculative and holding demand (arXiv:2211.07035)
• Absence of tangible value, price converges to liquidity value (arXiv:1802.09959)
• Both trading demand and speculative demand are significant for prices (GARCH) (arXiv:1812.09452)
• Supply-demand fundamentals and investor attractiveness impact prices (arXiv:1405.4498)
• Marginal production cost underpins value (refuting the worthless theory) (arXiv:1805.07610)
• Mining costs lag rather than lead prices (arXiv:2204.13102)
• Dual-value theory of artistic value and use value (arXiv:1904.05028)
• Scarcity, utility, trust, and the "greater fool theory" (arXiv:2010.01337)
• Beyond fundamentals: speculative and macroeconomic drivers (arXiv:1707.01284)
• Price models based on conviction and sentiment indicators (arXiv:1702.00215, arXiv:1709.08621)

Digital Gold and Store of Value

• Bitcoin = gold, Litecoin = silver (on-chain PU ratio) (arXiv:2308.00013)
• High volatility and high predictability, unsuitable as currency (arXiv:2103.00395)
• Bitcoin as a candidate sound currency (arXiv:2212.13986)

Bubble Dynamics

• Multi-scale bubble history, LPPLS (3 large bubbles and 10 small bubbles) (arXiv:1804.06261)
• Metcalfe's law + LPPLS, early warning of bubble collapse risks (timing is unpredictable) (arXiv:1803.05663)
• Real-time LPPLS collapse detection (arXiv:1905.09647)
• Application of social bubble hypothesis: author's claim that a hierarchy of bubbles brought Bitcoin into existence (arXiv:2005.06464)
• Self-sustaining bubbles, adaptive behavior (arXiv:2012.14860)
• Network hubs trigger bubbles and systemic risk (arXiv:1805.04460)
• Wealth effect of crypto bubbles, reminiscent of the 1929 real estate bubble (arXiv:2209.04385)

Volatility

• Daily and annual volatility are remarkably high (historical, forecasted, and implied) (arXiv:2401.02049)
• An extremely hazardous market entangled with volatility and jumps (arXiv:1912.05228)
• Rough volatility (Hurst < 1/2) (arXiv:1904.12346, arXiv:2511.03314)
• Roughness estimates are below 1/2 but are not invariant across time points and measurement designs (some samples are multifractal) (arXiv:2507.00575)
• Volatility asymmetry and time-varying efficiency (arXiv:2102.07425)
• Futures introduction more than doubled volatility (arXiv:2109.15052)
• Stabilization over time following the introduction of futures (arXiv:1906.03430)

Market Efficiency

• Mostly efficient, arbitrage opportunities under specific conditions (EMH) (arXiv:2504.18982)
• Time-varying efficiency (adaptive market hypothesis) (arXiv:1904.09403)
• Dynamic efficiency clustering, the coming of age of the market (arXiv:1901.04967)
• Excess returns possible by exploiting inefficiencies (arXiv:1805.08550)

Inflation Hedge and Safe Haven

• Bitcoin does not hedge inflation (arXiv:2301.10117)
• 2021 commentary: argument that it is none of currency, store of value, inflation hedge, safe haven, or tail hedge (arXiv:2106.14204)
• Weak and time-varying safe haven (arXiv:1703.00308)
• Weak safe haven and hedge (tail independence) (arXiv:1707.07977)

Stablecoins

• Stability is a fragile emergent state (SoK) (arXiv:2506.17622)
• Design vulnerable to speculative attacks on 1:1 peg (other classification) (arXiv:1905.11905)
• Trade-offs in price window design (inheriting reserve asset volatility) (arXiv:2602.15981)
• Fiat collateral as an anchor of stability / algorithms as risk amplifiers (arXiv:2602.18820)
• Stablecoins as dry gunpowder, risk contagion (arXiv:2603.23480)
• Central bank anchor mitigates liquidity vulnerability (SVB and USDC) (arXiv:2510.10469)