The persistent-transitory representation for earnings processes

Abstract

We consider the decomposition of shocks to a dynamic process into a persistent and a transitory component. Without additional assumptions (such as zero correlation) the decomposition of shocks into a persistent and transitory component is indeterminate. The assumption that is conventional in the earnings literature is that there is no correlation. The Beveridge–Nelson decomposition that is widely used in time series analysis assumes a perfect correlation. Without restrictions on the correlation, the persistent‐transitory decomposition is only set‐identified. For reasonable autoregressive moving average (ARMA) parameters the bounds for widely used objects of interest are very wide. We illustrate that these disquieting findings are of considerable practical importance, using a sample of male workers drawn from the Panel Study of Income Dynamics (PSID).

An impossibility theorem for wealth in heterogeneous-agent models with limited heterogeneity

“…where ξ t and η t are the persistent and transitory components, respectively (Ejrnaes and Browning, 2014). In this case y t = exp(ξ t + η t ) is a function of a Markov process z t = (ξ t , η t ), but y t itself may not be Markovian.…”

Estimation for Dynamic Panel Data With Individual Effects

“…Fractional modeling could be applied to errors of static models, in place of the popular ARMA modeling. In particular, issues of identification and estimation might be studied when the shocks to the errors comprise both a permanent and transitory component, as in e.g., Ejrnaes and Browning (2014). 7.…”

“…Fractional modeling of panel data has also been studied by Hassler, Demetrescu, and Tarcolea (2011). Fractional models might also be used in place of AR and the ARMA ones used by, e.g., Ejrnaes and Browning (2014), for income dynamics.…”

Identification of Joint Distributions in Dependent Factor Models

“…Bonhomme and Robin (2010) show identification of a model with mutually independent unobserved income shocks; and we show that the model is identified even when fixed effects and transitory income shocks are allowed to be arbitrarily dependent. Browning, Ejrnaes, and Alvarez (2010) and Ejrnaes and Browning (2014) contend that these models should have more unobserved heterogeneity and that income shocks are correlated. For example, transitory income shocks may represent bonuses, which could be dependent across time periods.…”