"Are Consumers Forward Looking? Evidence from Fiscal Experiments."

Abstract

How changes in current and future income affect consumer spending is a perennial question in macroeconomics and public finance. Theoretical constructs such as the permanent income hypothesis are of limited assistance in resolving this issue, because they neglect borrowing constraints and other market imperfections that can significantly affect the marginal propensity to spend out of current income. Recent empirical work has also proven inconclusive, since whether consumption responds to income fluctuations by more or less than the permanent income hypothesis predicts turns critically upon whether disposable income follows a random walk or is stationary around a longrun trend (see John Campbell and Angus Deaton, 1987). This controversy is unlikely to be resolved conclusively, because it is notoriously difficult to distinguish a stationary but highly persistent time-series from one that is nonstationary. These difficulties suggest the importance of searching for natural experiments, income shocks with predictable and well-understood effects on future income, to test models of consumer behavior. Changes in federal tax and transfer policy during the last two decades provide several episodes of this type.' Analyzing the effects of these tax changes is also central to understanding the role of fiscal policy in affecting national saving. This paper examines two aspects of consumer response to tax changes. Section I argues that consumption responds to temporary income tax shocks by more than the permanent income hypothesis would suggest. Results from the 1975 tax rebate in particular suggest that a $1 increase in transitory income raises spending by about 20 cents. Section II examines consumption responses to tax announcements. Although the results are not conclusive, they suggest that some consumers do not adjust consumption in anticipation of tax changes. The final section sketches some implications for analyzing fiscal policy.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 151 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.