External determinants of small business survival – The overwhelming impact of GDP and other environmental factors and a new proposed framework

This article claims that the basic relationships between, on the one side, the small firms’ hazard of exit, and, on the other side, the GDP growth rate and the industry growth rate are U-shaped. This means that there are more births and deaths for this segment of manufacturing enterprises during both cyclical downturns and booms in the economy. Higher competition in the economy in the first case comes from necessity entrepreneurs and in the second case from opportunity ones. The article also claims that the quadratic specification would rarely be the most adequate, since other combinations of different pairs of exponents would certainly better capture nuances of the relationships being regressed, in view of the fact that the actual U-shaped relationship is rarely symmetric. This is exactly why the artificial exclusive monotonic fitting normally produces parameter estimates that signalize the existence of a decreasing relationship. So, what may wonder many people, the invariably detected inverse relationship is not caused by the second half segment (where the economic upturns occur) of the continuous of the GDP growth rate, but rather by the first (where the economic downturns occur), whose impact on the hazard of exit is normally stronger. Also, even a direct relationship may occur because of this asymmetry and findings of lack of statistical significance result from a misguided attempt to fit a linear specification to a perfect, or almost perfect, symmetrical actual U-shaped relationship. The article conclusively claims that these realizations, and the fact that authors overfit by specifying contemporaneously the GDP growth rate, the industry growth rate and the industry entry rate, explain findings in the extant literature that are awkward, unexpected and embarrassing and interpretations that are many times completely inapplicable.