Criticisms of Bitcoin

This site's editorial policy rests on primary sources and neutrality. Understanding a technology requires more than listing its advantages. You need to know, with equal precision, what is not working, where the structural limits lie, and who objects on what grounds.

This topic presents the major criticisms of Bitcoin in their strongest form. We do not soften them, and we do not neutralize them preemptively with rebuttals. What central banks, academic researchers, and prominent economists actually wrote is set out here with dates and sources.

The scope is deliberately bounded. Individual misconceptions (the Ponzi claim, the quantum-computing threat, and so on) are examined in "Bitcoin Myths & Facts"; the histories of specific events such as Mt. Gox and FTX belong to "Incidents & Turning Points"; governance splits such as the block size war belong to "Forks & Chain Divergence". Here we treat the systematic critique itself.

This page is educational material, not investment advice. We make no price predictions whatsoever, and we mark clearly the line between what has been established and what remains contested.

Economics assigns money three functions: a medium of exchange (usable for payment), a unit of account (a denominator for pricing), and a store of value (preserving purchasing power). The most basic criticism of Bitcoin is that the first two barely function.

The difficulty as a unit of account stems from price movement. The all-time high set on 6 October 2025 was roughly $126,198; as of August 2026 BTC trades around $64,000, roughly 50% below that peak. Continuously denominating everyday prices in such a unit is, at least for now, impractical.

As a medium of exchange, one nation-scale experiment exists. El Salvador made Bitcoin legal tender in September 2021, but research by Alvarez, Argente and Van Patten (NBER Working Paper 29968, 2022) reported that although the law obliged every firm to accept it, only about 20% actually did, and only 11.4% reported any bitcoin-denominated sales at all. Across all sales covered by the survey, 4.9% were paid in bitcoin through the government's Chivo wallet, concentrated mainly among large firms; of the firms that did record bitcoin sales, 88% converted the proceeds into dollars. (The figures come from a face-to-face survey conducted in February 2022.) The paper concluded that usage was "low, concentrated, and has been decreasing over time."

On 29 January 2025 the Salvadoran legislature amended the Bitcoin Law by 55 votes to 2. The amendment made acceptance voluntary for businesses, required taxes to be paid in dollars, and removed Bitcoin's designation as "currency" (moneda), meeting conditions of the $1.4 billion IMF program agreed in December 2024. Voluntary private use remains legal, but critics cite this as a case where even state compulsion failed to establish Bitcoin as a medium of exchange.

Supporters answer that Layer 2 systems such as the Lightning Network carry the payments role, and that Bitcoin's primary use has shifted to store of value. Against that sits the counter-question of whether the goal was rewritten because the original one was not met. Which reading holds up better remains an open question.

The Bank for International Settlements devoted Chapter V of its June 2018 Annual Economic Report, "Cryptocurrencies: looking beyond the hype," to the argument that it is hard to identify a specific economic problem cryptocurrencies currently solve. Three points carried the case: a design requiring every user to verify the entire transaction history cannot scale with demand; payment finality is weak and can be reversed by forks (the report cites the unintended March 2013 chain split); and decentralized consensus consumes vast amounts of energy. Its conclusion was that the technology is "a poor substitute for the solid institutional backing of money."

Ulrich Bindseil and Jürgen Schaaf of the European Central Bank argued in their blog post "Bitcoin's last stand" (30 November 2022) that Bitcoin has never been used to any significant extent for legal real-world transactions, that its valuation rests purely on speculation, and that its environmental footprint is unprecedented. In a follow-up on 22 February 2024, after U.S. spot ETF approval, they maintained that Bitcoin's fair value is still zero.

Paul Krugman, in a New York Times column on 31 July 2018, identified two problems. Transactions require supplying a full history and consuming computational resources, so they are not frictionless; and there is no backstop tethering the price to anything real. From the latter he concluded that value depends entirely on self-fulfilling expectations, so a collective loss of confidence could render it worthless.

Warren Buffett called Bitcoin "probably rat poison squared" in a CNBC interview on 5 May 2018, arguing that it produces nothing on its own. Around the Berkshire Hathaway annual meeting held the same day, Charlie Munger said of cryptocurrency trading, "To me, it's just dementia." Both critiques converge on a single question: what justifies holding an asset that generates no cash flow?

Taken in their strongest form, these arguments share a premise: that the value of money rests on an issuer's institutions, taxing power, and legal enforcement, and that an asset lacking all three cannot perform money's role. That premise is itself contested, as the next section discusses. But it is worth establishing that these critics are not speaking from ignorance.

Traditional asset valuation starts from cash flows. Equities are valued on future earnings, bonds on coupon payments, real estate on rent, each discounted to present value. Bitcoin pays neither dividends nor interest, so this standard framework, discounted cash flow analysis, cannot be applied directly. This is the academic core of the criticism.

The strongest form of the argument is the ECB position quoted above: with no cash flow and no productive use, fair value is zero, and any price above zero merely reflects speculative expectation. On this view, a rising price is not a refutation: by the definition of a bubble, rising prices are entirely consistent with the claim.

The common rebuttal is comparative: gold, art, and collectibles generate no cash flow either, yet have commanded market prices for millennia. On the practical side, CFA Institute published its cryptoasset valuation guide, "Valuation of Cryptoassets," in November 2023, presenting several approaches in parallel: total addressable market, stock-to-flow, Metcalfe's law, and cost of production. The significant detail is the guide's own statement that "no single valuation model or metric should be used in isolation."

The honest statement today is therefore neither "Bitcoin cannot be valued" nor "its value is zero," but rather that no agreed valuation method yet exists. How to justify the price of an asset without such an agreed method remains academically unresolved, and that is precisely where the investment risk resides. Rebuttals to the blunt claim that Bitcoin "has no intrinsic value" are handled in "Bitcoin Myths & Facts".

The most frequently cited empirical study is "Blockchain Analysis of the Bitcoin Market" by Igor Makarov and Antoinette Schoar (NBER Working Paper 29396, October 2021). Using data through late 2020, they estimated that the top 1,000 investors controlled roughly 3 million BTC and the top 10,000 investors roughly 5 million BTC. Against the circulating supply of the time (about 18.6 million BTC) that is roughly a quarter. Where the paper itself speaks of "roughly one-third of Bitcoin in circulation," that phrase refers to the roughly 5.5 million BTC held by intermediaries such as exchanges, not to these investor totals.

The authors qualify their own estimate. Clustering blockchain addresses into real entities is an inferential technique, and they could not rule out that several of the largest addresses belong to the same owner. Their own reading is that true concentration may therefore be understated.

The same paper also separates intermediaries such as exchanges (about 5.5 million BTC) from individual investors (about 8.5 million BTC) on the basis of transaction patterns, and the top-1,000 and top-10,000 figures above refer to the individual side. Reading them as contaminated by exchange customer balances would be a mistake.

Care is needed, however, when reading the general on-chain distribution statistics that make no such separation. Exchange and custodian addresses pool the assets of many customers, so a large address is not the same thing as a wealthy individual. Since 2024, with spot ETFs and listed-company treasuries, this effect has grown: some top addresses represent hundreds of thousands of beneficiaries. Conversely, a single holder spreading funds across many addresses makes the distribution look flatter than it is, so the error runs in both directions.

One point of interpretation cannot be set aside: Makarov and Schoar note that this concentration itself implies the majority of the gains from further adoption are likely to fall disproportionately to a small set of participants. Supply design bears on this too. Newly issued supply shrinks with each halving, to 3.125 BTC per block as of 2026, roughly 164,000 BTC a year, so over time buying from existing holders becomes effectively the only route in.

In response, it is fair to point out that holdings of fiat, equities, and gold are also highly concentrated, so the verdict depends on the chosen benchmark. For a technology promoted under the banner of financial democratization, examining the actual distribution is reasonable; which benchmark it should be measured against remains the open question.

The first thing to establish is that electricity estimates shift substantially with both method and reference period. Conflating the two leads to misreading how far apart the figures really are.

The Cambridge Centre for Alternative Finance's Cambridge Digital Mining Industry Report (published April 2025), based on a survey of 49 mining firms covering roughly 48% of global mining activity, put annual electricity consumption for 2024 at about 138 TWh, up roughly 17% year on year and roughly 0.54% of global electricity use, with greenhouse gas emissions at about 39.8 MtCO2e.

The 2026 figures are a different matter. Cambridge's own daily index, CBECI, put its best guess at about 141 TWh a year as of 15 August 2026 (lower bound about 76 TWh, upper bound about 247 TWh, model v1.8.0). Digiconomist, working backwards from miner revenue with an economic model, estimates about 204 TWh annually and about 114 Mt of CO2 equivalent (retrieved 16 August 2026; the site publishes no reference date). Compared at the same point in time, the two best guesses differ by more than 40%, and CBECI's own lower and upper bounds span a factor of more than three. That spread comes from methodology, not from a difference in reference period. CBECI also revises its historical series whenever its model changes, so any citation needs to carry both the model version and the date. The carbon figures use different metrics again, so a simple ratio between them is not meaningful.

The critical case organizes into three points. First, absolute magnitude: a single network draws power comparable to a mid-sized country. Second, justification of use: the energy per transaction processed is orders of magnitude above other payment systems. Third, opportunity cost: the same renewable generation could have been directed elsewhere.

The second of those, energy per transaction, attracts a methodological objection of its own. Mining's electricity draw is a function of the security budget (the BTC price and fee revenue), not of how many transactions fit in a block; hold electricity constant and the figure still swings as Layer 2 or batching change the denominator. It is a serviceable comparison metric, but it cannot be read literally as "the cost of processing one transaction."

The "sustainable energy share" that supporters cite also warrants care. In the CCAF survey, sustainable sources (42.6% renewables plus 9.8% nuclear) reached 52.4%, but natural gas at 38.2% was the single largest source. The figure rests on self-reported data from surveyed firms and may not represent the whole industry. Even within the same report, an IP-based model yields about 69.6 MtCO2e, far above the survey-based 39.8 MtCO2e.

Whether that energy is "wasted" is not a factual proposition but a value judgment, and that debate is handled in "Bitcoin Myths & Facts" and "Bitcoin Economics". What this section establishes is narrower: numbers vary by method even at the same point in time, and comparing figures from different years widens the apparent gap further, which is itself part of why the environmental argument keeps losing resolution.

BIS Bulletin No. 69, "Crypto shocks and retail losses" (20 February 2023, by Cornelli, Doerr, Frost and Gambacorta), analyzed usage data from more than 200 exchange apps across 95 countries between August 2015 and December 2022. Its conclusion was that in nearly all economies, a majority of crypto app users lost money on their Bitcoin holdings. It further found that new users clustered into rising markets, while the largest holders reduced positions ahead of the Terra/Luna and FTX collapses.

The structural fragility of leveraged trading is a recurring target. On 10 October 2025, the derivatives data service CoinGlass reported roughly $19.1 billion of leveraged positions force-liquidated within 24 hours, affecting more than 1.6 million traders. BTC fell as much as 14%, and total crypto market capitalization contracted by hundreds of billions of dollars. Liquidation totals of this kind are structurally undercounted, because major exchanges throttle the liquidation events they publish through their data feeds, so the reported figure should be read as a floor rather than a full measure. The way derivatives amplify spot price movements is a by-product that has nothing to do with the design of the settlement network itself.

There is a cultural criticism as well. When "number go up" discourse dominates and price appreciation becomes the point, technical debate and operational problems are pushed to the margins. Meme-driven communication, the influencer economy, and the spread of fraudulent projects are argued to be continuous with that culture.

Notably, this criticism also comes from inside the Bitcoin community. Self-criticism is common: that abandoning self-custody, concentrating funds on exchanges, and centering all discourse on price hollow out the original design goal of a censorship-resistant payment network.

"Banking the unbanked" is among the most repeated promises made on behalf of cryptocurrency. Yet the World Bank's Global Findex 2025, surveying about 145,000 adults across 141 economies, reports account ownership at 79%, up from 74% in 2021, driven by mobile money and conventional financial accounts, with cryptocurrency not identified as a principal factor. Financial inclusion advanced; the evidence so far is that Bitcoin was not what advanced it.

Remittance costs are a second claim requiring scrutiny. According to the World Bank's Remittance Prices Worldwide, the global average cost of sending $200 eased from 6.49% in Q1 2025 to about 6.2% in Q1 2026, but remains more than double the UN Sustainable Development Goal target of 3%. There is little statistical evidence that Bitcoin has displaced that structure.

El Salvador offers more specific figures. Remittances received through crypto wallets fell from about $85.5 million in 2024, roughly 1% of total remittances, to about $57.67 million in 2025, a drop of some 32.5%, or around 0.6%, per central bank (BCR) data. From January to April 2026 they rebounded to about $23.1 million, up 44.3% year on year, yet still accounted for only 0.70% of total remittances over that period. That these are the levels in the country with the strongest possible tailwind deserves to be taken seriously.

Everyday payments have likewise stalled. Base-layer fees, confirmation times, and price volatility all sit poorly with retail commerce. The Lightning Network substantially relieves those constraints, but published statistics on payment volume and user counts are sparse, creating a separate problem: outsiders find adoption hard to verify (see "Lightning Network Primer").

In fairness, many of these promises appear nowhere in the Bitcoin whitepaper; they were layered on later by advocates and venture-backed companies. What the protocol failed to achieve and what the marketing over-promised are worth separating. Even after that separation, the fact remains that seventeen years on, everyday payment use has not become mainstream.

Bitcoin has repeatedly experienced severe declines. The principal ones: roughly 93% from June to November 2011; roughly 86% from November 2013 to January 2015; roughly 84% from December 2017 to December 2018; and roughly 77% from about $69,000 in November 2021 to about $15,500 in November 2022. Most recently, against the all-time high of roughly $126,198 on 6 October 2025, BTC traded around $64,000 as of August 2026, roughly 50% below that peak.

Declines exceeding 30% have thus occurred in every market cycle to date. This is not an exception traceable to any single event but a property repeatedly observed in this asset class. This site makes no prediction about future prices or recovery timing. The above is a record of what happened, not a signal about what comes next.

Risk is not confined to price. Serious defects have been found in the protocol and its software as well (the 2010 integer overflow, the unintended chain split of 2013, and CVE-2018-17144 in 2018, among others), and dependencies outside the protocol, internet routing and the software supply chain among them, remain part of the attack surface. The technical details are covered in "Bitcoin Vulnerabilities."

After seventeen years of operation, some questions can be considered effectively settled. Predictions that Bitcoin would inevitably go to zero, or that governments could simply switch it off, have not been borne out so far. Equally, the supporters' prediction that it would come into everyday use as a major medium of exchange has not been achieved either. The settlements run in both directions; neither side has won outright.

The open questions are more fundamental. First, there is no agreed valuation method. Second, whether a fee market alone can fund the security budget as halvings continue to shrink the block subsidy remains untested.

Third, mining pool concentration. Tallies such as mempool.space show that, over the trailing seven days as of 16 August 2026, the top three pools mined roughly 61% of blocks and the top four around 70%, which is no improvement as a decentralization metric. Fourth, the final shape of national regulation. Each of these awaits another decade of evidence.

On this question Makarov and Schoar, cited above, provide the strongest evidence. They measured concentration among miners themselves rather than pools, reporting that the top 10% of miners control 90% of mining capacity and the top 0.1% (about 50 miners, or 55–60 once untracked pools are accounted for) control close to half. They further found the concentration to be counter-cyclical, rising when the price falls sharply and after halvings, and concluded that the risk of a 51% attack rises in precisely those periods (data through end-2020). The reassurance that "miners can switch pools at any time" does not answer the concentration of the miners behind those pools.

When evaluating criticism, the essential skill is distinguishing its quality. Criticisms resting on factual error, such as the claim that Bitcoin is an untraceable anonymous currency, are examined in "Bitcoin Myths & Facts". Criticisms whose methodology is published and independently checkable (the BIS on payment finality, Makarov and Schoar on concentration, the BIS bulletin on retail losses) deserve to be engaged directly, even by supporters.

Stated neutrally: Bitcoin is neither a finished solution nor an obvious fraud, but a large-scale social experiment that has been running for seventeen years. That experiment contains parts that have succeeded, parts that have clearly failed, and parts that cannot yet be judged. We encourage readers to consult the primary sources on both sides and form their own assessment.