A Conceptual Model for the Integrated Policy Framework

The Mundell-Fleming benchmark says a flexible exchange rate plus a standard interest-rate rule fully insulates a small open economy, and this paper asks what that prescription survives once the world’s actual imperfections are added. It builds a three-period small open economy – households, a government, tradable-goods firms, housing-sector firms, domestic banks, and international financial intermediaries partly owned by domestic households – and gives a constrained social planner with full commitment four instruments: the policy rate, taxes or subsidies on capital inflows, sterilized FX intervention, and macroprudential taxes on domestic bank lending to households and to housing firms. Each can be used ex ante, in period 0 before a shock, or ex post, in period 1 after one. Countries differ along seven characteristics – currency of trade invoicing, commodity export share, stock of debt, currency mismatch, external debt limit, depth of FX markets, and housing-sector debt limit – and are hit by six shocks: productivity, commodity prices, the world interest rate, the external debt limit, foreign risk appetite, and the housing debt limit. The frictions are deliberately layered. Export prices are sticky either in the producer’s currency (PCP) or in a dominant currency (DCP), motivated by the observation that “many emerging markets have dollar invoicing shares above 80 percent.” An occasionally-binding constraint caps domestic banks’ external debt at a fraction of the domestic tradable price, in the spirit of Mendoza (2010), Bianchi (2011) and Farhi and Werning (2016); another caps housing firms’ debt at a fraction of their land value, following Kiyotaki and Moore (1997); and asset-market segmentation following Gabaix and Maggiori (2015) means intermediaries have limited capacity to bear the country’s currency exposure, so uncovered interest parity fails – the paper’s “shallow FX markets.” Five externalities follow: the standard Keynesian aggregate demand externality, a terms-of-trade externality the authors deliberately downplay and sometimes parameterise away, a pecuniary aggregate demand externality from the banks’ constraint interacting with currency mismatch, a pecuniary production externality in housing, and a financial terms-of-trade externality that exists only when FX markets are shallow. The results are a mapping from shock-and-characteristic combinations to instrument settings rather than a set of point estimates: this is a conceptual model illustrated by simulations, and it reports directions, signs and comparisons across regimes rather than calibrated magnitudes. Flexible exchange rates remain optimal for a class of cases, including under DCP when there are no financial frictions – though the DCP economy then needs larger exchange rate movements to do the same stabilising work. Financial frictions are what break the benchmark: when future shocks can make the banks’ constraint bind, prudential capital controls are warranted in normal times; capital controls and macroprudential consumer taxes are perfect substitutes only when the macroprudential perimeter covers the whole economy; FX sales and looser inflow taxes both buy monetary autonomy after a foreign-appetite shock, and buy more of it used together; and housing and external constraints can each trigger the other, so ex-ante housing macroprudential taxes may be needed in anticipation of external as well as domestic shocks. Three broad principles close the analysis: instruments are not interchangeable and a newly available tool may simply be the wrong one; instruments affect multiple imperfections, so adding one can raise or lower the use of another; and there is no strict assignment of domestic tools to domestic shocks or external tools to external shocks.

Filiz Unsal | Macro Paper Warehouse

A Conceptual Model for the Integrated Policy Framework