Stable currency flows and foreign exchange market spillovers

by Iñaki Aldasoro, Paula Beltran, Federico Grinberg

Compilation | Chen Asian

The data are from 64 centralized exchanges (including Binance, Coinbase, Kraken, etc.). For each stable currency-statutory currency, daily collection prices, volume of transactions and inflows and outwards from the legal currency to the stable currency are collected. For the construction of monetary-level indicators, trade volume weighted average prices are used and the flow of all exchanges aggregated。

Core definition 1: Equivalent deviation

There are two ways in which a subject wants to buy a stable currency in its own currency: directly (a stable currency is purchased in a centralized exchange) and indirectly (a fixed currency is exchanged in the instant foreign exchange market in United States dollars and then in United States dollars). In a full market without friction, the cost of both routes should be equal. Equivalent is defined as the direct price divided by the indirect price multiplied by 100 per cent. When the margin is greater than 100 per cent, it is more expensive to buy directly, implying arbitrage opportunities (stable currency in United States dollars, sold in exchange for cost); and vice versa. The more the margin deviates from 100 per cent, the greater the arbitrage。

Core definition 2: Net inflow rate

The net inflow rate is defined as the inflow from the legal currency to the stable currency minus the reverse outflow, excluding the outstanding stock against the currency-stable currency for the preceding period. This indicator measures new net financing (as a percentage of stocks) from a given legal currency into a stable currency ecosystem, reflecting net demand pressures for stable currencies. The positive value represents a net transfer of funds from French to stable currencies。

Core definition 3: Override interest rate parity bias

Overriding interest rate parity is a core indicator of the cost of dollar financing in traditional foreign exchange markets. It is equal to (forward minus spot rate) divided by the spot rate, less the difference between the local currency and the United States dollar. When the deviation is negative, it means that synthetic dollar financing is more expensive than borrowing directly into the United States dollar, i.e. there is a “dollar premium”. The greater the negative value, the higher the cost of financing the dollar. This is a classic indicator of the friction of finance in traditional foreign exchange markets。

Fact 1:There is a marked and uneven parity between stable currencies and traditional foreign exchange markets. In the sample, the main highly liquid currencies (such as the United States dollar, the euro, the pound sterling) had a small parity margin (average of 0.05-0.3 per cent) and were closely distributed around 100 per cent, indicating relative arbitrage efficiency. For economies facing high inflation, exchange rate fluctuations or capital controls (e.g. Argentine peso, Nigerian Naira, Turkish lire), however, the parity margin averaged several percentage points (e.g. the Argentine peso averaged about 1.06 per cent, with a maximum of 14.8 per cent) and was volatile. This fact suggests that there is an ongoing arbitrage between encryption and traditional markets, and that friction size is linked to macroeconomic vulnerability。

Fact 2:Net inflows of stable currencies were highly correlated with parities, devaluations of the currency and the expansion of the United States dollar premium. A linear regression (to track the impact of 0 to 15 days after the shock) revealed: an increase in net inflows of stable currencies would immediately and significantly push up the parity margin (up on the day of the shock); then result in a devaluation of the currency in the traditional spot market (a peak after 1-3 days); and a reduction in the interest rate parity differential (i.e., an increase in the United States dollar premium) in the short-term (3 months) coverage, while the impact on the 12-month coverage rate would not be significant. This suggests that the balance sheet constraints of the arbitragers are more tight in the shorter term. These correlations constitute prima facie evidence of the impact of stable currencies on traditional markets, but cannot yet be interpreted as causes and consequences。

Fact 3:THIS RELATIONSHIP CANNOT BE INTERPRETED DIRECTLY AS A CAUSAL LINK. THERE IS A CLEAR PROBLEM OF REVERSE CAUSALITY AND OMISSION VARIABLES. FOR EXAMPLE, THE EXPECTED DEVALUATION OF THE CURRENCY MAY SIMULTANEOUSLY DRIVE CAPITAL FLIGHT INTO STABLE CURRENCIES (INCREASED NET INFLOWS) AND LEAD TO DEPRECIATION ITSELF. GLOBAL RISK MOOD CHANGES (SUCH AS THE RISE IN THE VIX INDEX) AND MISSING FACTORS SUCH AS LOCAL REGULATORY SHOCKS MAY ALSO AFFECT BOTH STABLE CURRENCY DEMAND AND TRADITIONAL EXCHANGE RATES. A MORE RIGOROUS CAUSAL IDENTIFICATION STRATEGY IS THEREFORE NEEDED。

A theoretical model has been developed to establish a causal identification framework。The core set is as follows:

Residents need two assets at a time: a stable currency (for chain dollar access) and a synthetic dollar (for hedge, trade finance) obtained through foreign exchange swaps. The two are incomplete alternatives and demand varies along the lines of a stable currency premium and a United States dollar premium, respectively. When the stable currency premium rises, the demand for the stable currency falls; when the dollar premium expands (the covered interest rate becomes more negative), the demand for the synthetic dollar declines。

2. Stable currency issuers provide a fully flexible supply at the United States dollar peg, i.e. the dollar price of the Stable currency remains at US$ 1。

United States investors supply the dollar to the swap market, increasing as the United States dollar premium expands (the higher the premium, the greater the supply)。

A representative intermediary active in both stable and swap markets has three cost functions: a swap-specific cost (proportional to the square of the swing position), a stabilization-specific cost (proportional to the square of the stable position), and a cross-market cost (proportional to the square of the sum of two positions). The cross-market cost parameters are key and represent the additional marginal costs associated with an intermediary holding an overall currency opening. When the parameter is larger than zero, an increase in the stable position increases the marginal costs offered by the swap period, thus generating a spillover from stable currency markets to covered interest rate parity。

Model balance decomposition shows:

• Stabilizing currency demand shocks increases the stable currency parity differential (proportional effects)。

• If the cross-market cost parameter is greater than zero, the spread of interest rate parity (i.e., increases in the United States dollar premium and cross-market spillovers) is also expanded。

• The devaluation of the currency through a substitution effect (as the currency becomes more expensive or as the United States dollar premium rises, the population moves towards the immediate purchase of the dollar)。

When the cross-market cost parameter is zero, the two markets are completely decoupled, and a stable currency shock does not affect the coverage of interest rate parity and the United States dollar premium。

IN THE MULTI-COUNTRY EXPANSION, THE MODEL INTRODUCES “CROSS-BOOK PARTICIPANTS” — WHO ARE ALSO ACTIVE IN MULTIPLE STATUTORY-STABILIZED CURRENCY TRANSACTIONS, AND FACE TOTAL CONVERSION BUDGET CONSTRAINTS (THE SUM OF ALL CURRENCIES CONVERTED IS FIXED VALUES). WHEN THE URGENCY OF CONVERSION IN COUNTRY A ROSE, PARTICIPANTS WITHDREW THEIR ACTIVITIES FROM COUNTRY B, LEADING TO A DECLINE IN NET INFLOWS TO COUNTRY B, THUS AFFECTING PRICES AND EXCHANGE RATES IN COUNTRY B. THIS TRANSNATIONAL LINK PROVIDES THE THEORETICAL BASIS FOR THE INSTRUMENTAL VARIABLES FOR CAUSAL RECOGNITION。

4.1 Identification of challenges

There was a serious bias in the general minimum double-time return, as local demand shocks for stable currencies were highly correlated with other factors affecting local exchange rates (e.g. expected devaluation, changes in local interest rates, banking system pressures). There is a need for external sources of variability。

4.2 Precision tool variable construction

Step 1: Deprivation of net inflows through factor models Apart from global commons (e.g., global risk preferences, bitcoin price trends, dollar index changes, etc.), heterogeneity components are obtained. This step is equivalent to a reduction of all currency-shared fluctuations from net inflows of each currency-stabilized currency。

Step 2: For a targeted currency and a stable currency, the instrument variable is weighted by the size of the heterogeneity shock of all other currencies, and the weight is the market share of the currencies in the stable currency market (e.g., trade volume share). By excluding the target currency ' s own shocks, the tool variable is not relevant to the local factors of the target currency (e.g. local monetary policy, changes in local capital controls); at the same time, positive shocks in other countries can attract conversion activities away from the target currency through cross-book participants ' budgetary constraints, thus reducing the net inflow of the target currency and meeting the conditions of relevance。

4.3 Results estimated at a minimum of two times two stages

THE RETURN OF F STATISTICS DURING THE FIRST PHASE IS WELL ABOVE THE CONVENTIONAL THRESHOLD (OVER 500), INDICATING THE STRENGTH OF THE TOOL VARIABLE。

Estimated results for phase II (shock response, 1 per cent increase in net stable currency inflows):

• Price differentials: expansion of about 40 basis points with effects lasting about 10 days after the shock。

• Immediate exchange rate (in local currency): about five basis points of devaluation, with effects declining after several days。

• Three-month coverage of interest rate parity: a decline of about 5 to 10 basis points (i.e., an increase in the United States dollar premium), with significant effects and concentration in the short term。

• 12-month coverage of interest rate parities: no significant change。

Key rates: The absolute value of the covered interest rate parity factor divided by the parity margin is approximately 0.44, or about 44 per cent of stable currency price pressures, transmitted to traditional dollar financing markets. This median between 0 (full division) and 1 (full integration) indicates a significant but incomplete correlation between stable currency markets and traditional markets。

5.1 Cross-market costs are key drivers

Through model calibration (using literature and estimations of the above-mentioned tool variables), the counterfact analysis shows that:

Double the cross-market cost parameters and almost double the spread of the covered interest rate parity (from 6 to -12 basis points) and the depreciation of the exchange rate from 6 to 9 basis points。

• If cross-market cost parameters are close to zero (completely divided), cover interest rate parity premiums disappear and exchange rate depreciation halves。

• Stable currency-specific costs mainly affect the parities themselves and have limited impact on the spreads of covered interest rates。

Policy implications: In capital-account-opening economies, intermediaries can effectively reduce cross-market costs through international operations to hedge local currency exposures, while in highly regulated economies, regulation increases cross-market costs by discouraging the international decentralization of intermediary balance sheets, thereby enhancing spillover effects。

5.2 Dynamic expansion and dependency

The paper further built dynamic models and introduced:

:: Persistence of stable currency demand shocks (set at a self-return factor of 0.8, i.e. a half-life of about 3 days for shocks)

:: Intermediary wealth affects risk-taking capacity (the less the current wealth, the higher the unit position costs)

The simulation results show that:

• The shock has led to a reduction of about 5 per cent in the wealth of intermediaries, reduced risk tolerance and increased subsequent price responses。

• The effect of the same shocks is doubled when the initial wealth of the intermediary is below 50 per cent of the level of stability。

Accumulated pulse responses show that static analysis underestimates real spillover costs 5-6 times。

5.3 Foreclosure friction and crowding magnification

When a foreclosure mechanism is introduced (in the event of large ransoms, each unit of stable currency can only recover the value of less than US$ 1 as a result of a fire-line sale and settlement friction), a simultaneous flow shock with a friction shock (e.g. Terra/Luna collapse or USDC anchorage triggered by a Silicon Valley bank incident) could result in a spillover effect of five times the benchmark. This explains why the stabilization of currency pressure events have super-conventional effects on traditional markets。

For the first time, this paper demonstrates a causal spillover between stable currency markets and traditional foreign exchange markets。The main conclusions are as follows:

1. A 1 per cent increase in net inflows of stable currencies would expand the currency-French parity bias by about 40 basis points, the currency would depreciate by about 5 basis points and the short-term United States dollar premium would increase by 5 to 10 basis points。

2. About 44 per cent of stable currency price pressures are transmitted to traditional dollar financing markets, indicating a significant but incomplete correlation between the two markets。

3. Cross-market friction is the main determinant of the size of spills, and it is magnified by capital controls。

The spillover effect is state-dependent: the same shocks have a greater impact when intermediate capital is depleted; and foreclosure and flow shocks are multiplied several times。

Policy recommendations:

Precautionary requirements for stabilizing currency intermediaries: capital buffers, reserve liquidity requirements, centralized currency exposure limits, and possible reduction of spills from source。

2. Incorporation of macroprudential regulation: Policymakers (especially in emerging markets) should integrate stable currency market monitoring into the regulatory framework for foreign exchange and capital flows, and monitoring changes in spill rates could serve as an early warning indicator of increased connectivity. As stable currency markets continue to expand and mature, the same liquidity shocks may generate greater price responses in the future and spill-over risks may increase as markets develop. This discovery has far-reaching implications for understanding capital flows and financial stability in the digital age。