📈 economics

The Anchor Shift: Reconstructing Global Exchange Rate Determination—The China-U.S. Yield Spread as a Regime Variable, with Systematic Alternatives to Interest Rate Parity, the Mundell-Fleming Model, and the Triffin Dilemma

This paper proposes and validates a "Dual-Anchor Exchange Rate Theory" demonstrating that the global exchange rate system has shifted from a U.S. dollar single anchor to a China-U.S. dual anchor driven by their yield spread, a finding that systematically invalidates traditional models like Interest Rate Parity and the Mundell-Fleming framework while establishing a new unified pricing mechanism across global assets.

Original authors: Shuiping Tang

Published 2026-07-14

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Shuiping Tang

Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine the global money market as a giant, chaotic game of "Follow the Leader." For decades, everyone believed there was only one leader: the U.S. Dollar. The old rulebook, called "Interest Rate Parity," said that if you wanted to know where a currency was going, you just looked at the difference in interest rates between countries. It was like a simple math equation: High interest rates here, low interest rates there? The currency with the high rates should go down.

But this paper, written by an independent researcher named Shuiping Tang, says that rulebook is basically a blank page when it comes to the Chinese and U.S. relationship. In fact, the paper argues that the old math is so broken for the Chinese Yuan (RMB) that it explains almost nothing. When they tested the old theory against real data, the "explanatory power" (a fancy way of saying "how much of the story it tells") was a tiny, almost invisible 0.0001. That's not just a weak link; it's a broken chain.

The New Game: A Two-Anchor System

Instead of one leader, the paper suggests the world has shifted to a "Dual-Anchor" system. Think of it like a tug-of-war where the rope isn't tied to a single tree, but to two different posts: one in the U.S. and one in China.

Here is the surprising twist: The Chinese interest rate isn't just watching the game; it's actually pulling the rope for other countries!

The Chinese Anchor: The paper found that Chinese interest rates are the main boss for European currencies (like the Euro) and "commodity" currencies (like the Australian Dollar). When the Chinese interest rate moves, these currencies move with it. The paper shows that adding the Chinese interest rate to the mix boosts the ability to predict the Australian Dollar's value by a massive 0.38 (an R-squared value of 0.38), which is huge compared to the old methods.
The American Anchor: The U.S. interest rate still rules the Japanese Yen, but it's losing its grip on everyone else.
The Result: The global system isn't a tree where the U.S. Dollar branches out to everyone else. It's more like a parallel highway where the Australian Dollar and the Euro are driving straight toward the Chinese post, while the Yen drives toward the American post. There is almost no "traffic" (transmission chain) connecting the different currencies to each other anymore.

The Magic Switch: The Zero Line

How do we know which anchor is in charge? The paper found a "magic switch" located right around a spread of zero.

When the difference between Chinese and U.S. interest rates is positive, the U.S. anchor pulls harder.
When it dips near zero or goes negative, the Chinese anchor takes over.
The paper tested this with a "threshold regression" (a statistical test that looks for a tipping point) and found the switch happens right at 0.0078 (about 0.78%), with a statistical certainty so high the "p-value" is 0.000. This means the switch isn't a guess; it's a hard, measurable line in the sand.

Why the Old Rules Don't Work Anymore

The paper explicitly argues against several famous economic ideas:

Interest Rate Parity is Dead: The idea that interest rate spreads determine exchange rates is shown to be systematically wrong for the China-U.S. pair.
The "Single Dollar" Myth: The U.S. Dollar Index (DXY) is a terrible predictor for currencies like the Australian Dollar. The paper shows the DXY has almost no power over the Aussie dollar (an R-squared of just 0.01), while the new "Dual-Anchor" model is 51 times more powerful.
Capital Controls are Speed Bumps, Not Walls: Traditional theory says if a country controls its money (like China does), outside forces can't touch it. This paper says that's wrong. Chinese interest rates penetrate those controls and still price global currencies.
No "Natural" Interest Rate: The idea that interest rates always try to find a "fair" or "natural" value is rejected. Instead, the paper suggests rates just "wander" within wide intervals depending on the current economic regime.

The "Unified Field" of Spreads

The paper connects this to a bigger picture involving gold, oil, and bonds. It proposes a "Unified Field of Spreads," where the same China-U.S. interest rate difference acts as a master thermostat for the whole world.

Gold: The spread explains 55% of gold's price movements.
Euro: It explains 21%.
Australian Dollar: It explains 12%.
Chinese Bonds: It explains only 2%.

This creates a "pricing gradient." The spread is the boss of the whole system, but it influences different assets differently. The paper ran this through nine different types of rigorous tests (including "exhaustive testing" of 36 different factors and "LASSO" machine learning screening) to make sure the results weren't just a fluke. The results held up every time.

The Bottom Line

The paper concludes that we have moved from a world with one anchor (the U.S. Dollar) to a world with two (China and the U.S.). The old rules of economics, like the Taylor Rule or the Mundell-Fleming model, need to be rewritten because they assume a single anchor and a return to a "steady state" that no longer exists. Instead, the global economy is in a state of "regime switching," where the rules change completely depending on which side of the zero line the interest rate spread falls.

In short: The old map is wrong. The new map has two centers of gravity, and the line between them is right around zero.

Technical Summary: The Anchor Shift – Reconstructing Global Exchange Rate Determination

Problem Statement
This paper addresses a systemic dislocation in global asset pricing observed since 2022, where traditional international finance theories fail to explain exchange rate movements. Specifically, the core proposition of Interest Rate Parity (IRP)—that yield spreads determine exchange rates—has systematically failed for the China-U.S. exchange rate, with the explanatory power ( $R^2$ ) of the China-U.S. yield spread for monthly RMB movements dropping to near zero (0.0001). Furthermore, the prevailing "U.S. Dollar Single-Anchor" system (DXY) and models like Mundell-Fleming and Triffin Dilemma appear insufficient to explain the current pricing dynamics of global currencies, particularly the divergence between Chinese and U.S. interest rates and their impact on non-U.S. currencies. The paper seeks to determine if a new regime variable exists that can explain this dislocation and reconstruct the framework for global exchange rate determination.

Methodology
The study employs a rigorous, multi-layered empirical framework designed to eliminate factor mining and overfitting, inheriting and upgrading the methodology from Tang (2026a, 2026b, 2026c).

Data: Daily frequency data (February 2006 – April 2026) covering 3,973 trading days. Sources include FRED, ChinaBond, and the U.S. Treasury.
Factor Pool: A candidate pool of 36 factors, including Chinese and U.S. interest rate levels (various tenors), yield spreads (China-U.S., internal, and cross-country), and global factors (VIX, DXY).
Core Screening Mechanism:
1. Exhaustive Testing: Traversal of all single, two, and three-factor combinations to identify optimal models.
2. VIF Hard Constraint: A strict Variance Inflation Factor (VIF < 5) constraint applied to ensure factor independence, significantly stricter than the standard VIF < 10.
3. Machine Learning & Statistical Validation: LASSO regularization for independent factor selection, 200-iteration permutation tests to rule out random chance, and Newey-West corrections for autocorrelation and heteroskedasticity.
Robustness Checks: The study utilizes nine layers of validation, including multi-frequency Granger causality tests, 200-iteration Bootstrap threshold regressions, rolling window stability tests (40 windows), and pseudo-out-of-sample forecasting.
Theoretical Dialogue: The empirical results are systematically compared against classical theories (IRP, Mundell-Fleming, Triffin, Taylor Rule, DSGE, Fama-French, Behavioral Finance, MMT, and Portfolio Theory) to propose systematic alternatives.

Key Results

Systematic Failure of Interest Rate Parity: The China-U.S. yield spread levels show negligible explanatory power for RMB exchange rate movements across all tenors and frequencies (monthly $R^2$ = 0.0001). The spread functions not as a short-term trading signal but as a macro "regime variable."
Dual-Anchor Division of Labor: A clear structural shift has occurred from a U.S. single-anchor system to a China-U.S. dual-anchor system.
- Chinese Interest Rate Dominance: The Chinese interest rate (specifically the 2-year and 5-year rates) is the optimal pricing factor for European currencies (EUR, GBP) and commodity currencies (AUD). After controlling for local rates and traditional spreads, the incremental pricing power ( $\Delta R^2$ ) of the Chinese rate is substantial (0.20–0.38).
- U.S. Interest Rate Dominance: The U.S. interest rate remains the dominant anchor only for the Japanese Yen (JPY).
- Independence: The Chinese interest rate acts as an independent pricing dimension, penetrating capital controls to price global exchange rates.
Parallel Radiation Structure: There is almost no Granger-causal transmission chain between exchange rates (e.g., USD does not transmit to EUR, which then transmits to AUD). Instead, the global system operates as a parallel structure where currencies are directly priced by their respective anchors (Chinese or U.S.).
Regime Switching Threshold: Threshold regression identifies a critical switching point near a China-U.S. yield spread of zero (Bootstrap $p$ = 0.000). When the spread is positive, the U.S. anchor dominates; when negative, the Chinese anchor gains independent marginal pricing power. This threshold is consistent with findings in gold and crude oil markets.
Unified Field of Spreads: The same China-U.S. yield spread variable exhibits a systematic pricing gradient across five core global assets: Gold ( $R^2$ = 0.55) > Euro ( $R^2$ = 0.21) > AUD ( $R^2$ = 0.12) > Chinese Gov Bonds ( $R^2$ = 0.02) > RMB ( $R^2$ = 0.03).
Model Performance: The Dual-Anchor model (e.g., Chinese 2Y rate + VIX) outperforms the DXY single-factor model by a factor of 5.4 to 51 times in explaining exchange rate variance (e.g., AUD/USD $R^2$ jumps from 0.01 to 0.51).

Significance and Theoretical Contributions
The paper claims to offer systematic alternatives to foundational theories in international finance and macroeconomics:

Interest Rate Parity & DXY System: It proposes the "Dual-Anchor Exchange Rate Theory" as a replacement, arguing that the single-dollar anchor is obsolete for pricing non-Japanese currencies.
Mundell-Fleming Model: It revises the model via the "Dual-Anchor Penetration Hypothesis," demonstrating that Chinese interest rates penetrate capital controls to price global currencies, rendering capital controls a "speed bump" rather than a "wall."
Triffin Dilemma: It reframes the dilemma using a "Dual-Anchor Reserve Competition Model," where the yield spread acts as a quantitative thermometer for reserve currency competition, signaling a transition from unipolar to bipolar structures.
Taylor Rule & DSGE Models: It negates the existence of a stable "natural rate of interest" ( $r^*$ ), proposing an "Interval Wandering Hypothesis" where rates wander within macro-regime intervals. It further argues that DSGE models fail because pricing relationships exhibit structural breakpoints and do not return to a pre-shock steady state.
Factor Models & Portfolio Theory: It supplements the Fama-French model by showing factor exposure coefficients are regime-dependent (varying by a factor of 7 across regimes) and amends Portfolio Theory by demonstrating that asset correlations undergo regime switching near the zero-spread threshold.
Unified Field of Spreads: Together with three preceding papers, this work completes a theoretical loop, establishing the "Dual-Anchor Regime" as the unified pricing logic for commodities, fixed income, and currencies.

Conclusion
The paper concludes that the global exchange rate determination system has structurally shifted. The China-U.S. yield spread is the central regime variable driving this new "Dual-Anchor" system. The findings, validated through nine independent robustness checks and cross-validated by multiple methods, suggest that the era of the U.S. dollar single-anchor and the stability assumptions of classical models have ended, replaced by a regime-dependent, dual-anchor pricing framework.

Technical Summary: The Anchor Shift – Reconstructing Global Exchange Rate Determination

More like this