The first difference to institutional variables is negatively correlated to economic growth, indicating that people see institutional change as uncertainty. The relevant part of the large literature on the effect of uncertainty on growth is surveyed. The transition represents the necessary institutional changes that cause the smallest costs possible. However, institutional changes are often much larger than necessary. Thus, they are much more harmful to economic development than they need to be. The orders of magnitudes are assessed. It appears that the excess changes in institutions in many countries contribute considerably to the rareness of miracle growth.

The transition of the growth rate differs from the other transitions, as it has a first difference variable on the vertical axis. Hence, it is very fuzzy, but as the full sample consists of 10,329 observations, a clear pattern is found. It is hump-shaped, so that middle-income countries catch up with the high-income countries. However, the speed of the catch-up is slow. The hump-shape it rather robust. The hump-shape implies that low-income countries diverge and high-income countries converge, just as they do in all seven institutional indices. The hump-shaped transition curve is inconsistent with the one-sector Solow model that has been the workhorse model in the literature over the last 30 years. However, it is easy to explain by the two-sector models of development that used to be common. Two-sector models also show why miracle growth is possible, though difficult to reach.