This article examines the relationship between the competitive environment facing firms and the extent to which prices charged by these firms are responsive to changes in costs or demand. A model of optimal dynamic pricing is developed and tested on disaggregated industry data from five countries. The general conclusion is that in those sectors where competitive pressures - as proxied by several different indicators - are relatively weak, price smoothing is an important phenomenon. The implications of this finding for macroeconomic behaviour of economies are briefly considered, and in particular the light it sheds on "neoclassical" versus "fix-price" paradigms for macroeconomic theory.
Our tests emphasize implic'ations of sig- naling that do not appear to be particularly model-specific. Consequently, it may be possible to adapt our approach in order to test signaling models in other contexts. For example, one might test the hypothesis that education functions as a signal of ability by identifying observable employee character- istics that are systematically correlated with the costs of obtaining education. Under the signaling hypothesis, the returns to additional education should be higher for those individuals whose observable costs of educa- tion are also high.
This paper reviews the methods used for estimating potential output in OECD countries and the use of the resulting output gaps for the calculation of structural budget balances. The "split time trend" method for estimating trend output that was previously used for calculating structural budget balances is compared with two alternative methods, smoothing real GDP using a Hodrick Prescott filter and estimating potential output using a production function approach. It is concluded that the production function approach for estimating potential output provides the best method for estimating output gaps and for calculating structural budget balances.
Second, the coefficients of ADIV x HIRATED are all negative. In Table 1, however, the estimated values of these co- efficients are not statistically significant at conventional levels of confidence. More sig- nificant effects-both economically and sta- tistically-are found in Table 2. This is to be expected, since all bond ratings prior to 1977 are imputed. The effect of a high bond rating on the bang-for-the-buck is both neg- ative and statistically significant (at conven- tional levels) in equation (i) of Table 2. Statistical significance is somewhat lower in equation (ii). Since (as argued in Section 11) low bond ratings are probably associated with higher costs of paying dividends, a neg- ative relationship between bond rating and bang-for-the-buck corroborates the dividend signaling hypothesis and contradicts the alternatives.
This extend consumption smoothing intuition to stochasticcase: increase consumption by annuity value of present discountedvalue of expected additional income associated with current shock.
Fixing exchange rates between countries entering into the European Monetary Union shifts the burden of inter-regional adjustment onto labour and product markets. At present, rigid labour markets in European economies raise the output and employment costs of aqusting relative price levels at fixed exchange parities. High adjustment costs also reduce financial markets' confidence that fixed-exchange rates and fiscal convergence commitments are feasible, entailing large interest differentials. By contrast, the growing integration of Member countries should diffuse country-specific demand shocks more smoothly than at present, and reduce the impact of localised fiscal policy as well.
Barro’s tax smoothing hypothesis (TSH) implies that the government runs a ‘budget deficit’ whenever it anticipates the growth rate of national income to increase or the growth rate of its expenditure to decline. We test this implication of the hypothesis by examining the implied cross-equation restrictions on a vector autoregression (VAR) model using US. data for the period ranging from 1929 to 1988. Our formal tests reject the hypothesis for the full sample period, but cannot reject it for the post-1947 period. Further investigations show that the statistical rejection should be attributed to sharp differences in the statistical properties of the pre-1947 and the post-1947 data rather than the failure of the hypothesis itself. Key words: Government expenditures; Tax smoothing hypothesis; Budget deficits 1.
Fortunately, it is possible to test the hy- potheses of interest without attaching sig- nificance to particular values of THETA. Although the dividend tax rate does vary from year to year, much of this variation results from a small number of significant tax reforms. For example, although THETA rose from 0.691 in 1979 to 0.784 in 1985, more than half of this increase occurred between 1981 and 1982, when the first round of the Reagan administration's tax reforms went into effect. A second round of tax reforms increased THETA from 0.783 in 1986, to 0.825 in 1987, and to 0.880 in 1988. Regardless of whether one credits the par- ticular numbers, it is clear that the relative tax burden on dividends declined significantly after each round of tax reform. Con- sequently, we divided our sample period into three regimes: pre-1982 (regime I, high dividend taxation), 1982-1986 (regime 11, medium dividend taxation), and 1987-1988 (regime 111, low dividend taxation). We reestimated our basic specification for each regime, omitting THETA and interactions involving THETA. We then compared esti- mates across regimes. The signaling hypoth- esis implies that the bang-for-the-buck should decline from regime I to regime 11, and from regime I1 to regime 111. It also implies that the bang-for-the-buck should be lower for firms with higher bond ratings, and higher with greater capacity utilization, within each regime. The alternative hypotheses have the opposite implications.
The objection stated in the preceding paragraph suggests a powerful strategy for testing the hypothesis that changes in the bang-for-the-buck through time were, at least in part, attributable to changes in divi- dend taxation. Recall once again that the tax-uolicv regime shifts of 1982 and 1987 led to swift,-dramatic changes in effective divi- dend tax rates, and that these regime shifts accounted for much of the temporal varia- tion in THETA,. If the findings presented in Tables 1-3 are not spurious, then the process generating dividend announcement effects should have changed suddenly in 1982, remained relatively stable from 1982 to 1986, and changed suddenly once again in 1986. Moreover, it would be difficult to reconcile this pattern with any plausible al- ternative hypothesis. Certainly, sudden changes in the process governing dividend announcement effects would tend to rule out any explanation related to economic conditions that were trending more or less smoothly through time. Moreover, if the decline in the bang-for-the-buck was, for some reason, related to the other major financial market developments of this pe- riod (rather than to tax rates), then changes in the bang-for-the-buck should have been synchronized with these developments, rather than with tax reforms. It is therefore noteworthy that the acceleration of merger, acquisition, and repurchase activity was concentrated in 1984 and 1985, which fall squarely in the middle of a stable tax regime (see Bagwell and John B. Shoven, 1989).
In this figure, the sampling period is divided evolution of the excess-return relationship into the three tax regimes. The second is driven by spurious factors, one would not regime is further subdivided into sub-expect to obtain the same pattern of results, regimes IIa and IIb. We propose perform-except by some improbable coincidence. If, ing three tests for structural change across for example, the process of change was relaregimes and subregimes. Test 1 examines tively smooth, one would expect to obtain whether the relationship governing excess roughly similar test statistics for tests 1, 2, returns is the same in regime I as in regime and 3. Each joint hypothesis might or might IIa. Test 2 examines the stability of the not be rejected; the key point is that there excess-return relationships for subregimes would not be a dramatic. svstematic differ