[ Pobierz całość w formacie PDF ]
.360).Applying this ratio, the targetadjustment for Japan would amount to a total of about 1.9 percent of GDP,or 0.63 percent annually over three years.This direct effect is substantial,gauged against a baseline growth rate of 1.5 to 2 percent.ConclusionAlthough the principle of cyclical adjustment suggests that some delaywould be appropriate, eventually Japan should likely not be exempt fromthe international decrease in current account surpluses (or increase indeficits) that will be required as the counterpart of US reduction of itsoutsized current account deficit.The range of targets considered in thispaper for the US adjustment and Japan s share suggests that in a relativelyfavorable outcome (such as that in variant B of table 9.6), by 2005 Japancould reduce its current account surplus by perhaps some $40 billion(compared to the current pace of $120 billion annual surplus) as its sharein a $280 billion US current account deficit reduction.This would involveachieving domestic growth on the order of about 2.5 to 3 percent annuallyover this period, or 1 percent above baseline.This outcome would require policies capable not only of securing theextra growth above baseline but also of compensating for about 0.6 percentof GDP annually that would be lost in real demand from the decline inreal net exports associated with the adjustment.It would also likely requirea real effective exchange rate appreciation of about 12 percent, which inturn would likely imply a real appreciation against the dollar of about21 percent.This would take the yen from its current level of about 122per dollar to 101 per dollar, which is also about the level that is consistentwith the historical trend in the real value of the yen over the past twodecades.The policy implications of these findings would seem to include thefollowing.First, once Japan begins to show sustained recovery, it wouldseem inappropriate for Japanese (or international) authorities to seek toblock the gradual appreciation of the currency toward this range.23 Second,23.Note that, in contrast, Ito (2002) identifies 115 yen per dollar as a revealed interventionrate at which the Bank of Japan has sought to halt the rise of the yen in recent years.THE IMPACT OF U.S.EXTERNAL ADJUSTMENT ON JAPAN 197Copyright 2003 Institute for International Economics | http://www.iie.com the need for the United States to curb its external deficit and for Japansooner or later to absorb some portion of the counterpart surplus reductionwill likely be an additional reason beyond the already challengingdomestic difficulties, including banking sector fragility for the Japaneseauthorities to pursue aggressive measures for economic stimulus for sometime.Their task is a daunting one for which there are no easy recipes, inview of the Japanese economy s unusual combination of circumstancesrecession and deflation despite near-zero interest rates; high governmentdebt ratio; banking sector balance sheet problems; and a weak stockmarket.Appendix 9.1Modeling Japan s Current Account BalanceIn a previous publication (Cline 1995) I set forth a forecasting model forJapan s current account balance, centered on the following relationship:pf(1) ln zt ln [ ]ln R* [ (1 ) ]ln RLtLtpdln Yd ln Yf [ ]ln Rtwhere zt is the ratio of nonoil imports of goods and services to exportsof goods and services; pf is the foreign price of the imported goods inforeign currency; pd is the domestic price of the exported goods in domesticcurrency; is the absolute value of the price elasticity of imports; isthe price elasticity ( 0) of exports; R is the nominal effective exchangerate; R* is the real effective exchange rate; subscript t refers to the currentperiod; operator L in the subscript refers to a weighted average of theprevious eight quarters; is the income elasticity of imports; is theelasticity of exports with respect to foreign income; is the pass-throughparameter from the exchange rate to import prices; and is the pass-through parameter from the exchange rate to export prices.24 When pass-through from exchange rate change to trade price change is complete,the pass-through parameters take the values 1 and 0.Ifinstead pass-through is only 85 percent (for example), then 0.85and 0.15.This appendix reports new estimates of this model, using quarterlydata from 1980 through the second quarter of 2000.The InternationalMonetary Fund s index of nominal effective exchange rate (NEER, basedon relative unit labor values) is used for R; its corresponding index for24.See Cline (1995, 14).Note that the lag operator has the following weights on priorquarters beginning with the quarter prior to the present: 0.067, 0.117, 0.15, 0.167, 0.167, 0.15,0.117, 0.067.198 DOLLAR OVERVALUATION AND THE WORLD ECONOMYCopyright 2003 Institute for International Economics | http://www.iie.com the real effective exchange rate (REER) is used for R*; and pf /pd is calculatedas NEER/REER (Cline 1995, 20).Weighted foreign GDP growth is basedon the IMF s reported quarterly real GDP data for Japan s eight largesttrading partners, weighted by 1990-95 shares in Japan s exports.25 Japan squarterly real GDP is also as reported by the IMF (2002a).Quarterlyimports and exports of goods and nonfactor services are from the samesource.The series on oil imports refers to Standard International TradeClassification (SITC) 33, and data are from the Ministry of Public Manage-ment (2002).The regression estimated is:pf(1 ) ln zt ln 1 ln R* 2 ln RLt 3 ln Yd 4 YfLtpd5 ln Rt diDiiwhere 1 [ ], 2 [ (1 ) ], 3 , 4 , 5[ ], and Di are dummy variables for the quarter in question.Theequation is estimated using Cochran-Orcutt correction for autocorrelation.The resulting estimated parameters are as follows, with t-statistic in paren-theses:6.28 ( 6.9);1 2.056 (7.8);2 1.471 ( 5.1);3 1.297 (4.4);4 0.331 ( 2.4);5 0.256 ( 3.2);d1 0.0324 (2.2);d2 0.0128 (0.8);d3 0.0132 ( 0.9);rho 0.242;Adjusted R2 0.5971.Figure 9A
[ Pobierz całość w formacie PDF ]