Hall’s findings are that it is impossible to predict life-cycle consumption into the far future. After a few quarterly predictions, consumption should be treated as an exogenous variable. Even with predictions of one or two quarters, any prediction is as good as it gets and can only be slightly improved with current stock prices.
The utility function the marginal utility is assumed to be a linear function of consumption as this implies that consumption follows a random walk apart from a trend. Functional form wise this makes sense as the greater the consumption the greater the utility that should be received, however by now we are aware of diminishing returns and as such should take this with a grain of salt and perhaps a assumption that takes that into account would be a better assumption.
Et u’(ct+1) = [(1 + delta)/(1 + r)] u’(ct)
The data that hall uses is the consumption of nondurables and services in 1972 dollars from the U.S. National Income and Product Accounts divided by the population.
Hall finds that lagged wealth reduces the standard error of the consumption regression function by 20 cents. It is consistent with the permanent income hypothesis, as the hypothesis infers that wealth in previous quarters should have no predictive value on consumption for this quarter, which the results confirm.
Fiscal policy which partly concerned with taxation can have an impact on consumption. These policies must be new otherwise they are already taken into account and they must have an impact on permanent income, like taxes.