The determination of wages and prices constitutes an important field of applied econometrics as well as a very relevant topic in that context of changing economic conditions we have been experiencing since the beginning of the present decade. Econometric literature in that field is assuredly vast and sophisticated, but the tracking ability and the predictive quality of these studies are by now somewhat doubtful, for most of them embody two serious flaws: 1) their sample periods don't cover many quarters in the seventies, 2) most of the time series privileged one independent variable (labor market pressure variable or inflationary expectations variable). We legitimately questioned the relevance of such a choice in a recessionary-inflationary context the origins of which are manifold. Subsequently, we decided to increase the number of observations covering recent quarters and to expand the set of explanatory variables. After having provided the reader with a formal justification of our choice of independent variables, we ran two sets of regressions. One with SHAZAM and one with FEDEASYSHAZAM diagnosed quite well a bunch of difficulties: autocorrelation, multicollinearity, and a limited number of degrees of freedom impairing likelihood tests. We tried to fix them up with FEDEASY. We selected an adequate procedure: a joint combination of a COCHRANE-ORCUTT iterative procedure and ALMON lags. This technique yielding results purged from autocorrelation multicollinearity, heteroskedasticity, and increasing the number of degrees of freedom. Once these technical problems are solved, we can still explain a substantial part of the variance of our dependent variables in five regressions out of ten ran with FEDEASY. The main finding of our investigation is the existence of an intermediate-run positively-sloped inflation-unemployment relationship for the U. S. Economy. Ultimately, we briefly laid down the microeconomic foundations of our econometric results.