Duke Model: $8B 51% Bitcoin Attack Could Be Profitable

Campbell Harvey, a finance professor at Duke University, presented a model that estimates a 51% attack on bitcoin could cost about $8 billion and be profitable if attackers combine majority hashpower with large short derivatives positions. He discussed the analysis during a podcast interview.

Under the scenario Harvey described, a well-funded group would quietly assemble enough mining hardware and infrastructure to control more than half of the network’s computing power while simultaneously building substantial short positions in bitcoin on derivatives markets. The attackers would then use the network advantage to reorganize blocks, censor transactions or double-spend funds and aim to drive the price down so the short positions gain value.

A 51% attack gives control of the longest valid chain to the party with majority hashpower and lets that party influence which transaction history nodes accept. It does not allow attackers to create bitcoin without valid signatures or to take funds directly from properly signed wallets. It can, however, disrupt transaction finality and enable repeated transactions on a private chain to double-spend.

Harvey said the presence of liquid derivatives changes the economic calculation that has long made such an attack appear irrational. “The difference today is the derivatives markets,” he said, adding that offshore venues allow traders to establish leveraged short positions that profit from price declines. He estimated the cost at about “50 basis points of the value of bitcoin” under his assumptions.

The $8 billion estimate depends on hardware prices, energy costs, total network hashrate and how long the takeover must run. Building the required capacity would require large numbers of application-specific integrated circuits, data center space and long-term power contracts. Such a buildup could leave observable traces in semiconductor orders, construction of mining facilities, power agreements or unusual derivatives activity.

Defensive options exist outside the protocol’s longest-chain rule. Exchanges could limit or scrutinize suspicious derivative positions, miners could redirect computing power, and software developers and users could coordinate updates or reject an attacker’s chain. Any coordinated response would be disruptive and politically contentious and could be difficult to organize quickly.

The podcast host questioned whether a buildup on the scale Harvey describes could be assembled without detection and whether an attack would push bitcoin’s price low enough for shorts to recover billions in costs. The host also noted that other proof-of-work networks have experienced and survived 51% attacks.

Harvey described the scenario as theoretical and did not predict an imminent attack. He said the potential for a derivatives-backed motive is a risk factor that investors and network participants may want to consider.