Trilemma, Dilemma, or Amalgamation? The Effect of Global Forces on Domestic Interest Rates

In a landmark 2015 paper provocatively titled “Dilemma Not Trilemma”, Hélène Rey argued that the policy space for small open economies does not, after all, afford a choice, given open capital accounts, between exchange rate stabilization and monetary policy independence. The bulk of the paper documents a global financial cycle under which capital flows, credit, leverage, and asset prices all move in sync across international boundaries. It is then taken on faith that with such ebbs and flows of finance washing around, the only way to carve out space for independent monetary policy is through capital controls or, in a more evolved spirit, macroprudential measures. Any empirical examination of the interplay between the exchange rate and monetary policy under the classic trilemma was beyond the scope of the paper.

A perspective on the global financial cycle from the domestic side of a small open economy was presented by Assistant Governor of the Bangko Sentral ng Pilipinas Margarita Debuque-Gonzales at our ACAES session at the 2026 Allied Social Science Association Annual Meeting. Debuque-Gonzales identified periods of disconnect between short-run market rates of interest and the central bank policy rate in association with foreign capital flows into the Philippines. Applying a Markov-switching model, she showed that under global risk-on conditions, inflows of foreign capital drove the Philippine treasury-bill rate down below the policy rate to undermine monetary policy transmission. With the switch to risk-off conditions, short-term rates reconnected to the policy rate. The key insight is that applicability of the trilemma is state dependent.  

Attempting to formalize the effect of external shocks on monetary policy, analysts have incorporated an exchange rate term into a Taylor-type rule. Taylor (2001) himself surveyed research in this vein early on and concluded that policy-rate reaction to exchange rate movement stood to achieve little if any improvement in macroeconomic outcomes, and indeed could make things worse. A more recent study by Lu, Xia, and Zhou (2022) considered the case of China where, in contrast to the floating rate environment assumed by Taylor, exchange rate stabilization is intrinsic to the policy mandate. Markov switching was again applied, in this case to capture three states of central bank intervention in response to external shocks: strong forex purchase; strong forex sale; and weak market intervention in either direction. The exchange rate then entered the policy rule directly for each state. The conclusion was that policy attention to exchange rate stabilization significantly constrains central bank response to domestic imbalances, China's capital controls notwithstanding.

The problem with models that simply add an exchange rate term to a Taylor rule to discern the impact of external shocks on policy is that exchange rate movement is an incomplete measure of exchange market pressure when central banks intervene, as they do routinely in emerging economies. A three-state Markov switching formulation of intervention only begins to address the problem. A proper rendering of forex market pressure would need to capture central bank intervention and exchange rate movement in combination. A standard measure of exchange market pressure dating to Girton and Roper (1977) involves equal weighting of the two arguments. But this formulation is arbitrary and not very convincing. For purposes of assessing the consequences of external shocks for domestic stability and monetary policy, it would be preferable to rely on empirical inference.

A graphical framework for capturing exchange market pressure (EMP) in two dimensional form is developed in my textbook, Macroeconomics for Emerging East Asia. We apply the framework in Figure 1 to exploring the transmission of external shocks to domestic interest rates for Indonesia and the Philippines. The horizontal axis measures appreciation of the local currency vs the US dollar. The vertical axis measures reserve flows on the balance of payments relative to reserve stocks. Positive EMP arises from a positive balance of payments shock, such as a capital inflow. This drives local currency appreciation that may be offset to a degree by central bank purchase of foreign exchange. We define the positive EMP zone as extending 30 degrees in either direction beyond the upper right quadrant to accommodate the central bank leaning slightly into appreciation (lower right quadrant) or more fulsomely into depreciation (upper left quadrant) with pressure still positive on balance. Conversely, negative EMP results from a negative balance of payments shock, such as a capital outflow, where this brings a tendency for currency depreciation that the central bank may lean against through sale of reserves. Again the boundary is extended 30 degrees into adjacent quadrants to capture predominance of negative pressure. 

The hypothesis is that a positive balance of payments shock will tend to push market interest rates down. This holds whether inflowing funds are passed through, allowing the exchange rate to rise, or absorbed by the central bank, expanding its balance sheet and stimulating credit growth. Conversely, a negative shock will tend to put upward pressure on interest rates. Observations for which the interest rate moves opposite the sign of exchange market pressure, as expected, are shown in orange in Figure 1. Given the many other factors acting on interest rates, a fairly strong shock will presumably be needed for an external effect to prevail. We define an ellipse for each of the positive and negative zones to separate more extreme outliers from the pack, choosing width and heighth parameters to delineate differences as starkly as possible. With this approach, we find confirmation of the expected effect on interest rates of an external shock in either direction for both countries. Observations lying outside the ellipses are orange in 80 percent of cases or more vs 40-63 percent of cases inside the ellipses.

What does this mean for trilemma vs dilemma? It's complicated. Most of the time the economies examined operate inside the ellipses where external forces show no systematic effect on interest rates, the implication being that central banks maintain a grip on monetary policy. Under more extreme instances of external shock, interest rates do appear to be impacted. This is consistent with the finding of Debuque-Gonzales for the Philippines that during global risk-on periods when capital seeks out emerging markets, short-term interest rates drop below the policy rate for a debilitating impact on monetary policy effectiveness. Under these circumstances, central banks may find recourse in sterilizing their forex market interventions with a tightening effect that would bring market interest rates back into line with the policy rate. The downside of this strategy is that sustaining higher interest rates would fail to discourage continued capital inflows. If the expansionary effect of these inflows were triggering inflation, central banks would be in a bind. A trilemma choice choice between stabilizing the exchange rate and preserving internal balance might prove elusive. That leaves Rey's option from the erstwhile trilemma of imposing capital controls to preserve space for the other two legs of the triangle.

 __________________________

My thanks to the Bangko Sentral ng Pilipinas for support of this research.