Stable coins don’t inflate crypto markets
To what degree has aggregate stable coin issuance driven the price of Bitcoin and other cryptocurrencies? This column addresses the Griffin and Shams (2018) collusion hypothesis based on results in Lyons and Viswanath-Natraj (2019). Our bottom line: We find no systematic evidence of stable coin issuance driving cryptocurrency prices. We do find, in contrast, evidence of alternative hypotheses for the drivers of issuance. Specifically, (i) stable coin issuance endogenously responds to deviations of the secondary market rate from the pegged rate and (ii) stable coins perform a significant role in the digital-asset economy as a safe haven. This can be seen, for example, in the significant premiums during the COVID-19 panic of March 2020.
For perspective, stable coin use has risen dramatically in the last two years, with estimates of total trading between Bitcoin and Tether, the largest stable coin, exceeding the volume of Bitcoin/USD in 2019.1 That stable coin use should be growing so rapidly is consistent with their ‘raison d’etre’ – to solve the store-of-value problem by pegging their value to the US dollar (Yermack 2015).
If a stable coin is managed by a centralised issuer, then in principle coin supply can be increased at its discretion, leading to potentially inflationary effects on crypto asset pricing. This question is important since Tether Inc. – the company behind Tether, the largest stable coin by market capitalisation – is facing litigation. As of October 2019, there is a class action suit against Tether that claims (i) Tether is not backed one-for-one by dollar reserves and (ii) Bitfinex increased Tether issuance during 2017 to manipulate Bitcoin markets.2 Allegations assert collusion between Tether Inc. and Bitfinex to increase stable coin issuance to drive the rapid appreciation in Bitcoin prices that occurred in late 2017.3 The collusion view finds support in the recent Griffin and Shams (2018) paper, which provides evidence consistent with Bitfinex expanding the supply of Tether to fuel a boom in Bitcoin in late 2017.4 Alternatively, if stable coins are used purely to satisfy transaction demand by investors, we would expect no systematic effect of stable coin issuance on the price of Bitcoin.
A primer on Tether
To understand the role of Tether in potentially driving inflation of crypto asset prices, we first document how Tether is created. Figure 1 outlines that creation via collateralised dollar deposits.5
Figure 1 Tether creation process up to 2018
Analogous to a currency board, every Tether issued is in principle 100% backed by a dollar deposit, so that in the event of a run on the currency, all investors could redeem their Tether for an equivalent amount of dollars. Tether is created when an investor deposits dollars into Tether’s account, creating an equivalent supply of Tether that is introduced in circulation. Prior to 2018, nearly all the Tether created via grants was immediately distributed to Bitfinex and on to the other exchanges for trading in the secondary market.
A change occurred in 2018: Tether Treasury now retains a fraction of the total Tether in circulation. Figure 2 separates total Tether in circulation into the secondary market and Tether kept in the reserves of the Treasury, with the 2018 change clearly evident. This is important for our analysis: To understand the impact of Tether on Bitcoin, we need total Tether in circulation net of Tether stored as reserves by the Treasury.6 The Treasury, through its reserve holdings of Tether, has the capacity to sell Tether for dollars in the event the Tether price in the secondary market is priced above parity.7
Figure 2 Tether supply held by the Treasury and in the secondary market
Note: The figure plots total Tether in circulation, separating the amount in circulation in the secondary market (held by investors and exchanges) and the amount held by the Tether Treasury as reserves. Data are from Omniexplorer and Etherscan apis.
Our first test is to measure the effect of shocks to Tether supply on the price of Bitcoin, after controlling for past movements in supply and Bitcoin price. This can control for two-way feedback and for lagged effects, for example. Using our more precise measure of Tether flow to the secondary market, we find no significant effect on prices of major non-stable crypto currencies (Figure 3). This result is robust to the choice of sample period – including the late 2017 period in which Bitcoin prices surged – and holds for other major stable coins as well.8
Our results do not preclude the possibility that price manipulation has occurred; however, based on aggregate issuance data, there is no systematic effect. This supports the view that stable coins operate more as a decentralised system of exchange rate pegs, without first-order intervention effects on prices via issuance. Instead, we argue in our paper that stable coins’ use as a vehicle currency is dependent on other factors (Lyons and Viswanath-Natraj 2019). We now turn to these other factors.
Figure 3 Response of Bitcoin and Ethereum prices to Tether issuance
Note: This figure documents local projections of a change in secondary-market issuance on the prices for Bitcoin and Ethereum, using the method in Jordà (2005). Data for secondary market flows are from Omniexplorer and Etherscan. Price data are from cryptocompare. Controls include lagged price changes, and changes in cryptocurrency fundamentals hash rate and number of unique addresses of the network, based on Bhambhwani et al. (2019). Sample period is from August 2017 to November 2019. Gray area denotes two-standard-error bands for statistical significance at the five-percent level, indicating no significant price impact over the subsequent 20 days.
Fundamentals of stable coin issuance
One factor that theory tells us can drive flows of Tether into the secondary market is the incentive to arbitrage deviations of Tether’s market price from the peg. For example, consider the case where the dollar price of Tether in the market is above parity. In this instance, an investor can buy Tether from the Treasury at a one-for-one rate, and sell Tether at the prevailing market rate for a profit, resulting in a flow of Tether from the Treasury to the secondary market.9 Our empirical analysis finds strong evidence of this: a 100 basis point (1 cent) increase in the dollar price of Tether results in a flow of approximately $0.3 billion to the secondary market. Arbitrage by secondary-market participants offers a decentralised solution to exchange rate stability.
A second factor that theory tells us can drive Tether flows is the role of stable coins as a vehicle currency. In periods of risk, some investors will choose to exchange into a better store of value. Portfolio rebalancing toward Tether and other stable coins provide this function with minimal intermediation costs. On some exchanges, for example, there are long processing lags for dollar withdrawals in order to comply with intermediation procedures.10 During the period of collapse in Bitcoin markets in January and February of 2018, the price of Tether traded up to $1.05, a premium of 500 basis points (5 cents) in the secondary market, and data on trading in the Bitcoin-Tether market suggests a significant rebalancing of portfolios away from Bitcoin and towards Tether during this period. We also find quantitatively similar premiums in Tether and other stable coins during the COVID-19 panic of March 2020, in which the price of Bitcoin fell by 40% in a single day on 12 March. We test this more rigorously and find that returns on Tether and other major stable coins comove negatively and consistently with a cryptocurrency risk factor that explains the movements of cryptocurrency prices like Bitcoin and Ethereum. The negative correlations between stable coin returns and the risk factor are more pronounced during periods of increased volatility in Bitcoin markets, a fact also noted in Baur and Hoang (2019).
This column answers a series of questions relevant to whether stable coins have an inflationary effect on crypto asset prices. The bottom line: We find no systematic evidence that stable coin issuance affects cryptocurrency prices. Rather, our evidence supports alternative views, namely, that stable-coin issuance endogenously responds to deviations of the secondary market rate from the pegged rate, and stable coins consistently perform a safe-haven role in the digital economy.
Baur, D G and L T Hoang (2019), “A Crypto Safe Haven Against Bitcoin”, Available at SSRN.
Bhambhwani, S, S Delikouras and G M Korniotis (2019), “Do Fundamentals Drive Cryptocurrency Prices?”, Available at SSRN.
Griffin, J and A Shams (2018), “Is Bitcoin Really Untethered?”, Available at SSRN.
Jordà, Ò (2005), “Estimation and inference of impulse responses by local projections”, American Economic Review 95(1): 161–182.
Lyons, R K and G Viswanath-Natraj (2019), “What Keeps Stable Coins Stable?”, Available at SSRN.
Yermack, D (2015), “Is Bitcoin a real currency? An economic appraisal”, In D L K Chuen (ed.), Handbook of digital currency (pp. 31-43), Elsevier.
We thank the Berkeley Haas Blockchain Initiative for a research grant.
2 Plaintiff estimated damages amount to roughly $1.4 trillion.
3 For more on the suit against Tether, see The Economist
4 This note does not address their findings specifically. They depart from our analysis in fundamental ways, for example, in providing evidence that Bitfinex channelled Tether to an investor who is linked to approximately 50% of the net buying of Bitcoin during this period. The data we use in our analysis are aggregated and cannot therefore speak to their disaggregated results.
5 The schematic in Figure 1 shows creation of Tether in circulation through dollar deposits from Bitfinex, a cryptocurrency exchange in partnership with Tether. Starting in 2019, other market participants were able to deposit dollars directly with the Tether Treasury.
6 Some analyses of Tether’s impact on Bitcoin have missed this point.
7 As an analogy, the Tether Treasury’s holdings of Tether is similar the holding of foreign reserves by a central bank. If a currency pegged to the dollar is pegged at an overvalued rate, the central bank can intervene in the foreign exchange market by selling dollar foreign reserves and buying domestic currency.
8 See Lyons and Viswanath-Natraj (2019) for our full analysis. Our aggregate issuance data means we cannot trace movement of Tether from the Treasury to specific wallets precisely.
9 Conversely, when the dollar price of Tether is below one, an investor can buy Tether at the exchange and sell to the Tether Treasury, resulting in the opposite flow from the secondary market to the Tether Treasury.
10 The cryptocurrency exchange Bitfinex states that it will take investors 7 to 15 days to make dollar withdrawals from their platform. Also, fees are often imposed when dollar withdrawals are frequent or large.