Testing the permanent-income hypothesis: new evidence from West-German states (Länder)

New Evidence on the Excess Smoothness of Consumption

“…The evidence for US states [12] supports generally the permanent-income hypothesis. The evidence for West German states [9] supports the presence of excess smoothness. Overall the results are disparate and no specific conclusion can be made.…”

“…Tests of this hypothesis in the literature were conducted for the US [8,12,20,21], for the UK [23,24], for OECD countries [22], for UK regions [11], for US states [12], and for West German states [9]. There is no clear pattern in the findings.…”

“…Bagliano and Bertola [6,17] report an actual ratio of 0.64 for quarterly data, much higher than the one estimated with the data in this paper. Malley and Molana [8: Table 3, p. 1032, and [9] find statistically different theoretical ( 1 ) and empirical ( 0 ) impacts for Germany, with a range for 1 between 0.907 and 1.284, and a range for 0 between 0.093 and 0.351. These estimates for 0 are mostly higher than the empirical estimates of this paper which range between 0.0322 and 0.1129 (see column 6, Table 4).…”

“…*Address correspondence to this author at the Faculty of Business Administration & Economics, Haigazian University, Mexique Street, Kantari, Beirut, Lebanon; Tel/Fax: +961 1 349230; E-mail: samih.azar@haigazian.edu.lb If income follows an ARIMA(1,1,0) data generation process, with a partial autoregressive coefficient , then consumption ( C ) will react to the income innovation t +1 in the following manner [6][7][8][9][10][11]:…”

“…These three methods should yield the same size of the impact, and all must be equal if the RWPIH is true. Often the income process and the empirical regression of the income innovation on consumption are estimated separately, although some authors have estimated the model jointly, making it possible to test for the equality of the theoretical and empirical impacts, like [9] for example. Both procedures are followed in this paper.…”