Sources universitaires et de recherche
A Theory of the Consumption Function
Friedman's statement of the permanent-income hypothesis.
The PIH benchmark predicts a much larger response to a change perceived as permanent than to one perceived as transitory. Accessed 2026-05-23.
Princeton University Press · 1957
Stochastic Implications of the Life Cycle-Permanent Income Hypothesis
Hall's random-walk implication for consumption.
Robert Hall (1978) derives an unpredictable-consumption-change benchmark under rational expectations plus additional preference and interest-rate restrictions. Accessed 2026-05-23.
Journal of Political Economy · 1978
Household Expenditure and the Income Tax Rebates of 2001
Households spent 20 to 40 percent of their 2001 rebates on nondurable goods during the three-month period in which the rebates arrived. Accessed 2026-11-14.
Johnson, David S., Parker, Jonathan A., and Souleles, Nicholas S. · 2006
Consumer Spending and the Economic Stimulus Payments of 2008
Counting durable goods, mainly vehicles, households spent 50 to 90 percent of their 2008 stimulus payments. Accessed 2026-11-14.
Parker, Jonathan A., Souleles, Nicholas S., Johnson, David S., and McClelland, Robert · 2013
Consumption, Income, and Interest Rates: Reinterpreting the Time Series Evidence
Aggregate consumption data fit a model in which half of consumers follow the permanent-income hypothesis and half consume their current income. Accessed 2026-11-14.
Campbell, John Y., and Mankiw, N. Gregory · 1989
Fondements de recherche
Éléments probants et données
The simulator uses only current income, one lagged income observation, the current-income weight alpha, and the propensity k. Wealth, credit access, uncertainty, and expectations data are relevant to fuller PIH models but are not inputs here.
Calibrage ou mesure
Permanent income is the two-observation estimate Y_P = alpha Y + (1 - alpha)Y_previous, and planned consumption is C = kY_P. The residual Y - Y_P is mechanical, not a statistically identified transitory shock.
Limites
- The filter is backward-looking and does not solve lifetime optimization.
- No stochastic orthogonality, interest rate, wealth, borrowing constraint, or precautionary motive is imposed.
- Hall's martingale benchmark is cited as external context, not generated by this page.
Conseils d’utilisation
Quand le modèle suffit
Use the page to illustrate partial updating and the exact current-income response k times alpha in a two-observation consumption estimate. It is a transparent teaching approximation to permanent-income intuition.
Quand il reste une esquisse
Do not use it to identify permanent versus transitory income, estimate a structural MPC, evaluate a tax reform, or test Hall's random-walk implication. Those questions require a forward-looking model and an empirical measurement strategy.
Quand changer de modèle
Switch to a structural forward-looking life-cycle or permanent-income model when lifetime resources and expectations are the objects of interest. Use a buffer-stock model for precautionary saving and borrowing limits, or HANK when liquid-wealth heterogeneity determines the aggregate response.
Signal de réfutation
Given fixed alpha and k, planned consumption that does not move by k times alpha for a one-unit current-income change rejects this filter's calibration. A high measured MPC out of a transfer can reject a fuller unconstrained PIH benchmark, but it does not by itself falsify this two-observation estimate because alpha and k govern its response.