OPEN-ECONOMY MARKET: Model Calibration 3

w6675-19
In Figure 2, we see that a unit aggregate demand shock leads to an upward jump in output that tails away as time passes. This blip inyt leads to a real depreciation (i.e. qt rises), because of the increased import demand; via (5.15), the depreciation brings about a drop in уThe fall in inflation results from complicated dynamic expectational effects involving both price adjustment and consumption behavior; inflation would rise instead of fall if there was positive serial correlation in the v, process. The fall in inflation brings about a small temporary decline in Rt.
w6675-20
More interesting, perhaps, are the responses to an unexpected unit increase in real income abroad. The dominant effect, according to Figure 3, is a unit appreciation in the exchange rate, both real and nominal. This increases у, and>>, hand-in-hand, and leaves pt, Aph and Rt almost entirely unaffected. Figure 4 shows that a unit increase in у t — a favorable technology shock — also has no appreciable effect on pu &Ph and Rt but drives qt and st in the opposite direction from a yt* shock more.

Here the reason is that the increase iny, leads to a sizable (though smaller) increase in income (output), which involves an upward jump in import demand that can only be satisfied by a depreciation in the real (and nominal) exchange rate. The lack of gradual adjustment in these two figures is a consequence of the exact random-walk nature of die shocks. If stationary AR(1) processes were posited instead, all panels would look quite different.
w6675-21

Finally, we have in Figure 5 the responses to a shock to the “risk premium” term in the UIP relation. From (4.13) it is clear that an upward blip in к, will lead to a jump in the same direction in s{ (and, with sticky prices, in qt). The increase in qt leads to an expansion of export demand and therefore to an upward jump in output that wears off as time passes. But what is the explanation for the fall in pt shown in the middle left panel, in Figure 5?

In the period of the qt jump, there is no effect on pt since the latter is predetermined. Then there are in succeeding periods expected real exchange rate appreciations, which, by reducing import costs, lead to price level decreases. These wear off as time passes, as the value of qt returns to its initial level.
w6675-22
It will be clear from the foregoing discussion that many — perhaps most — of the responses of interest are intimately linked to the open-economy features of our model. If it were closed down, by assigning a value of zero to the parameters (£Y&yy&j) an(j the model would behave quite differently in several respects.

Tags: , ,