A long-standing tax deferral strategy, known as “buy, borrow, die,” is quietly introducing substantial hidden credit risk into decentralized finance (DeFi) lending pools. This traditional tactic, adopted by crypto investors, allows them to avoid immediate capital gains taxes by borrowing against appreciated digital assets rather than selling them outright. Lisa Simone research indicates this strategy is gaining traction.
However, new research from academics at the University of Texas at Austin and the National University of Singapore highlights that this tax-driven behavior can lead to borrowers becoming less reactive to declining collateral values, thereby increasing the likelihood of defaults within DeFi protocols.
Lisa Simone research on crypto’s allure
The “buy, borrow, die” strategy traditionally allows asset holders to maintain exposure to their investments while accessing liquidity without triggering taxable events. In the crypto sphere, an investor who bought Ethereum (ETH) at $1,000 and watched it rise to $4,000 might want to cash out $1,000. Selling a quarter of their ETH would incur a $750 capital gains tax under US rules.
DeFi presents an alternative. The investor can deposit their entire ETH holdings into a lending protocol, using it as collateral to borrow $1,000 in a stablecoin. This loan is not immediately taxable income, the investor retains full ETH exposure, and they gain spendable funds without ever selling their original asset.
Tax deferral versus immediate liquidation risk
While advantageous for tax planning, this approach creates a delicate financial balance within DeFi. A $1,000 debt secured by $4,000 in ETH starts with a manageable 25% loan-to-value (LTV) ratio. But a drop in ETH’s price to $2,000 instantly doubles that LTV to 50%, with accumulating interest pushing it even higher.
If the LTV surpasses the lending protocol’s predetermined limit, the collateral becomes subject to automatic liquidation. This process allows external traders to repay a portion of the loan and claim some of the pledged ETH at a discount, aiming to protect the lending pool.
Research reveals hidden risks in DeFi lending
Academics Lisa De Simone of the University of Texas at Austin, Peiyi Jin of the National University of Singapore, and Daniel Rabetti of NUS explored this intricate connection between tax planning and DeFi credit risk. Their working paper scrutinized the Venus protocol, a significant DeFi lending platform operating on the BNB Smart Chain.
The researchers analyzed approximately 13 million transactions from November 12, 2020, through July 31, 2022, covering the 15 largest tokens on Venus. Their dataset generated 1.36 million daily borrower observations, revealing that about 3% of traders experienced what the study defined as a default.
Defining default in decentralized finance
A DeFi default differs significantly from a traditional missed mortgage payment. The paper classified a borrower as defaulted when their loan remained above Venus’s 60% loan-to-value limit for at least seven consecutive days. Crucially, this had to occur without any subsequent borrowing or depositing activity from the user.
The study recorded a total of $133.34 million in outstanding defaulted debt, a figure representing accumulated daily exposure rather than unique principal losses. This means a single problematic loan could contribute to the default totals across multiple days, highlighting persistent risk.
US tax law alters borrower behavior
To isolate tax-motivated behavior from general market volatility, the researchers leveraged a key external event: the enactment of the US Infrastructure Investment and Jobs Act on November 15, 2021. Section 80603 of this law expanded information-reporting requirements for brokers dealing with digital assets.
This change prompted US traders to anticipate increased reporting of their crypto activities to the Internal Revenue Service (IRS). While the actual reporting on Form 1099-DA for custodial brokers began in January 2025, and basis reporting in January 2026, the law’s enactment in late 2021 created a perceived shift in visibility for US taxpayers, while international users remained unaffected.
The study inferred US users based on activity during US business hours, unusual behavior on US holidays, and holdings of dollar stablecoins under US oversight. This methodological approach allowed them to compare how likely US-linked borrowers were to trade assets compared to international users following the law’s enactment.
This type of regulatory shift can significantly impact how individuals manage their portfolios, often leading to adjustments in trading frequencies or asset allocation, a common trend seen across various financial markets.
Reduced trading activity and deferred gains
The findings were stark: US-linked borrowers became 24.5% less likely to trade assets compared to their international counterparts after the law was enacted. For borrowers utilizing stablecoin debt, this decline was even more pronounced, showing an additional 23% reduction in trading activity. This aligns with the “buy, borrow, die” incentive, where borrowers seek to retain appreciated collateral while using borrowed stablecoins for liquidity.
This pattern intensified among borrowers with larger paper gains and higher loan-to-value ratios. Trading activity notably fell in December, particularly in its final week—a common period for investors to defer gains into a new tax year. Activity subsequently increased once holdings passed the one-year mark, qualifying for lower US long-term capital gains rates.
The researchers estimate that US borrowers in their sample deferred an average of $3,357.42 in capital gains tax annually. This figure, representing approximately 17% of their trading portfolios during the study period, underscores the financial incentive driving this behavior. Such patterns underscore why long-term crypto holders often prefer a buy and hold strategy over frequent market timing.
The compounding impact on DeFi lending pools
The reduced trading activity directly translates into increased credit risk for DeFi lending pools. Borrowers disincentivized from realizing gains are more likely to leave risky accounts open longer, miss opportunities to repay debt, or fail to add collateral when LTV ratios deteriorate. This means a personal tax preference originating outside the protocol can directly influence the level of unresolved debt within it.
Using an instrumental-variable design, the study estimated that a 1% increase in tax-induced illiquidity was associated with an 11.2% rise in defaulted accounts and a substantial 39.6% increase in defaulted loan value. A one-standard-deviation increase in this illiquidity corresponded to approximately $350 more defaulted debt per borrower, a figure 2.7 times the baseline value in their model.