A Quantitative Microfounded Model for the Integrated Policy Framework

Many emerging market central banks that run inflation targeting frameworks nonetheless intervene in foreign exchange markets, and some use capital flow management tools, particularly when capital flows are volatile – an asymmetry with advanced economy central banks that this paper sets out to quantify rather than merely describe. The vehicle is a microfounded two-country New Keynesian model with two financial frictions similar to those in the three-period model of Basu et al. (2020): a Gabaix-Maggiori (2015) limit on the risk-bearing capacity of FX-market traders, which makes the uncovered interest parity premium fluctuate inefficiently, and an occasionally binding external debt limit in the sudden-stop tradition, which can produce sharp spread increases, current account reversals and deep contractions. To these are added conventional price and wage stickiness plus indexation and high exchange-rate pass-through to import prices, which stand in for weakly anchored medium-term inflation expectations, and a nonlinear Kimball aggregator that lets inflation respond asymmetrically – more to depreciations than to appreciations. The model is calibrated to a representative small open emerging market and, for comparison, a representative small open advanced economy; crucially, “we deliberately assume no systematic differences in the conduct of monetary policy between AEs and EMEs,” so the asymmetry has to come from structure. The calibration is disciplined against outside evidence: the FX-shallowness parameter is 0.06 for the emerging market and 0.02 for the advanced economy, with the emerging market value chosen so that FX purchases worth 10 percent of GDP produce roughly a 15 percent depreciation, in line with Adler, Lisack and Mano (2019); intermediary and portfolio-investor home-ownership shares of 0.75 deliver an unhedged FX exposure for the median emerging market of around 16 percent of GDP, in line with IMF (2021); and the debt limit is set so steady-state debt sits 12 percentage points of annual GDP below the constraint, putting the economy in the constrained regime about 3 percent of the time. Model impulse responses to a 10 percent depreciation sit inside the 90 percent confidence bands estimated by Brandao-Marques et al. (2021) for both country groups, and reproduce contractionary depreciations in emerging markets. The central experiment is a risk-appetite shock, AR(1) with persistence 0.95, scaled to depreciate the advanced economy’s real exchange rate by about 10 percent: in the advanced economy it behaves like an expansionary demand shock that policy can look through, while in the emerging market inflation rises persistently, the central bank tightens, and output contracts. The authors then evaluate simple rules using stochastic simulations, with welfare measured in permanent consumption-equivalent units and, for robustness, a quadratic loss with a weight of one-third on the output gap. An FX intervention rule leaning against the portfolio-driven UIP premium eliminates sudden stops entirely in their setup and improves welfare by more than simply removing the debt limit – so most of the benefit comes from smoothing the UIP premium, not from crisis prevention. A precautionary capital flow management rule leaning against net foreign liability accumulation also generates substantial gains, part of them from taxing foreign investors and from a stronger average real exchange rate. Combining the two does better still. But an FX intervention rule that fires only when the borrowing limit binds, while it does blunt the spread spike, prevents the needed external adjustment, and both welfare criteria judge it detrimental on balance. Finally, in an advanced-economy liquidity trap where output falls about 8 percent and the policy rate is pinned at zero, sterilized FX purchases of 20 percent of annual trend GDP in the first quarter stimulate output and keep CPI inflation closer to target – Svensson’s “foolproof way” – though when the same policy is run by a bloc half the size of the world economy and the foreign economy is also at its lower bound, the spillovers can be “sizable beggar-thy-neighbor effects, even to the extent of being globally contractionary in the short run.”

A Quantitative Model for the Integrated Policy Framework

Many emerging market economies moved off fixed exchange rates to inflation targeting over the past two decades, yet unlike advanced economies they kept intervening in foreign exchange markets and, in some cases, using capital flow management tools. This paper builds an empirically-oriented New Keynesian small open economy model – “similar to those widely used by central banks” – to quantify when that behaviour improves policy tradeoffs. It extends Galí and Monacelli (2005) along four dimensions: a broader set of real and nominal rigidities, including habit persistence, sticky wages and prices, and imperfect exchange rate passthrough with local currency pricing as the benchmark (producer and dominant currency pricing are also available); adaptive inflation expectations for some agents, capturing imperfect monetary policy credibility; incomplete markets, so domestic agents must borrow in a foreign currency bond rather than share risk fully; and Gabaix (2016) discounting in the Euler, UIP and price-setting equations, which mitigates the forward guidance puzzle. Onto this largely log-linear core the authors graft three nonlinearities: a UIP risk premium that rises sharply once net foreign liabilities pass a risk-tolerance-dependent threshold, a private borrowing spread that rises nonlinearly as the currency depreciates (following Bruno and Shin, 2018), and an effective lower bound on the policy rate. Calibration is quarterly and, apart from the balance sheet channels, identical across country types with one exception – the price and wage formation parameters, set to 0.75 structural persistence with steeper Phillips slopes for the emerging market against 0.5 and 0 with flatter slopes for the advanced economy. The linear results establish the tradeoff. A UIP risk premium shock (AR(1) of 0.9, scaled to depreciate the advanced economy’s real exchange rate by 10 percent) “looks very similar to a ‘standard’ aggregate demand shock” in the advanced economy, where policy can look through a transient inflation rise; in the emerging market the same shock raises inflation persistently, forcing real rates up and crowding out domestic demand, so output contracts slightly while inflation rises substantially. Policy cannot resolve this within the rate instrument alone: a more aggressive inflation-stabilisation rule limits the inflation rise but produces “a sharper output contraction of about 1.5 percent.” The nonlinear results establish where the extra tools help most. In a crisis scenario combining a risk tolerance decline (AR(1) of 0.85, half-life about five quarters) with falling foreign demand, a high-net-foreign-liability economy suffers a real depreciation several times larger and a considerably bigger output contraction than a low-liability one, and either FX sales or an outflow tax limits the depreciation, allows easier policy, and damps the borrowing spread – allaying what the authors call the “stagflationary” effect. But the relief is intertemporal: by supporting the currency, both tools slow the trade balance improvement, so net foreign liabilities end up higher and the UIP premium eventually runs above baseline, and the appeal “is somewhat diminished… as the shock becomes more protracted.” Stochastic simulations over 20,000 periods show the nonlinearities generating a left skew in domestic absorption reaching about -15 percent and in the output gap about -10 percent, and systematic FXI and CFM rules “markedly reduce the likelihood of large real exchange rate depreciations” and the associated tail risk of a large output contraction. A final exercise applies the linear advanced economy model to a liquidity trap where output falls over 10 percent below baseline and the policy rate is pinned at zero, and finds FX purchases keep core CPI inflation much closer to its 2 percent target – Svensson’s “foolproof way.” The authors are unusually direct about the limits of all of this: “our model results should not be taken as an unqualified endorsement of the use of these tools either in general or in specific situations,” since the model assumes perfect foresight while real decisions are made under uncertainty, and since stabilising the exchange rate “may impede the development of hedging markets, and encourage an excessive buildup of foreign currency debt.” A title-page disclaimer records that the document “was prepared before COVID-19 became a global pandemic and resulted in unprecedented economic strains” and “does not reflect the implications of these developments and related policy priorities” – worth carrying, since the paper’s policy discussion predates the episode that most tested its subject matter.

Christopher Erceg | Macro Paper Warehouse

A Quantitative Microfounded Model for the Integrated Policy Framework

A Quantitative Model for the Integrated Policy Framework