There is less clear evidence that having connections to the federal government is bad for growth. While these connections appear to lower growth in real personal income, there is no evidence that it lowers real annual GNP growth, and it somewhat counter-intuitively appears to increase investment.28 For example, estimates in columns 7–8 imply that a one standard deviation increase in spetsbanks' relationship to the federal government leads to between a 0.31 percentage point reduction and a 0.28 percentage point increase in the annual growth rate. However, there is much stronger evidence to suggest that spetsbanks most different from other banks are bad for growth; estimates in columns 9 and 10 indicate that a one standard deviation increase in the dissimilarity index results in annual growth that is between 0.27 and 0.60 percentage points lower, though neither is quite statistically significant at the 10 percent level.