In studying the movement in the production workers’ share in the value added (i.e., labor’s share), the following models were considered by Gujarati:*

Model A: Y_t = β₀ + β₁t + u_t

Model B: Y_t = α₀ + α₁t + α₂t₂ + u_t

where Y = labor’s share and t = time. Based on annual data for 1949–1964, the following results were obtained for the primary metal industry:

Model A: Ŷ_t = 0.4529 − 0.0041t                 R² = 0.5284             d = 0.8252

                                    (−3.9608)

Model B: Ŷ_t =0.4786 − 0.0127_t + 0.0005t²

                                    (−3.2724)     (2.7777)

                                                                R² = 0.6629                     d = 1.82

where the figures in the parentheses are t ratios.

a. Is there serial correlation in model A? In model B?

b. What accounts for the serial correlation?

c. How would you distinguish between “pure’’ autocorrelation and specification bias?