Some insight into the time series stability of the term structure of spreads at short maturities can be gleaned from Eurocurrency spreads, for which data over a longer time interval are more readily available. Lund’s model implies that the instantaneous spread is Markov, so the slope of the term structure of spreads contributes no additional information. A crude proxy for the former that partially avoids the impact on short maturities of speculative attacks is the six-month Eurocurrency spread; the latter is proxied by the difference between 12- and 6-month spreads. Table 2 reports estimates of the regression
for weekly data over 1977-1998, estimates for c over the 1990-98 subsample used by Lund, and heteroskedasticity-consistent Chow tests of process stability.
People need payday loans like speedy cash because they all need money at some point. If you arrived at this point, you probably need a loan with low APR and fair terms that you can be sure about. You are welcome to visit read to get a loan like this, and we can guarantee you will not be sorry about this decision.
With the exception of the Dutch-German spread slope, all subsample estimates of c1990^98 are statistically insignificant, justifying Lund’s specification. Over the longer 1977-98 interval, spread slopes are statistically significant for Denmark, Italy, and the Netherlands, but not for the other five countries. However, it does appear that including the spread slope captures parameter instabilities in the estimated spread process to some extent. P-values are substantially lower for subsample stability tests of Lund’s AR(1) than for stability tests of (26), for 7 out of 8 countries. Parameter stability of (26) is rejected at the 10% level for 3 countries (Belgium, Great Britain, and Sweden), whereas the subsample stability of Lund’s AR(1) is also rejected for these three, Denmark and Italy.
Prospects for the Future
There has been considerable speculation regarding what monetary policies will be pursued by the new European Central Bank (ECB). Some have argued that the ECB is likely to initially pursue a tight monetary policy, in order to establish an anti-inflationary reputation. The selection of Duisenberg as its initial head is certainly consistent with this perspective. Others have argued that the voting power of countries such as Italy that historically have been less concerned about inflation may result in monetary policy somewhat weaker than that historically pursued by the Bundesbank.
Morgan (1997) argues that the EMU probability calculators indicate a weak-ECB scenario. Fluctuations in the average EMU probability of three “periphery” countries (Italy, Spain, and Sweden) have been strongly correlated with deutsche mark weakness against the dollar. The argument is illustrated in Figure 4, which shows that the average Italian-Swedish 1999 forward spread against the DM has indeed fluctuated roughly in tandem with the $/DM exchange rate.
One difficulty with this argument is that forward spreads over this period are a good proxy for the overall level of European bond yields and forward rates, and for German rates in particular. Figure 5 shows German 3-month Eurodeposit rates and forward rates at various maturities over the same interval. Falling German rates were accompanied by even greater declines in non-German rates, while rising German rates accompanied larger increases elsewhere. The synchroneity was most pronounced for forward rates of at least 1 year in maturities.
Consequently, in terms of the implications for the $/DM exchange rate, it is impossible to distinguish the weak-ECB hypothesis from an alternate hypothesis that the exchange rate fluctuations reflected German forward interest rate fluctuations over 1992-96 resulting, e.g., from shifting near-term expectations of future Bundesbank monetary policy.