Interesting study from 2015 investigating 158 countries from 1960 to 2007 and comparing human development across electoral authoritarian regimes (these regimes have a legislature in which elections are held, but the executive branch is functionally completely independent and runs the show) and democracies using instrumental variables. Electoral authoritarian and democracies perform comparably in reducing infant mortality.
We see that they also perform similarly across other indicators of development:
The one area in which electoral authoritarian regimes performed worse than democracies on was when it came to political liberties, perhaps unsurprisingly (Table 3). Since they provide both the coefficient and the t-statistic in their tables, it's possible to manually compute the CIs of the effect sizes.
Civil Liberties
EA History: Coefficient = 0.15, t = 2.33, SE = 0.15/2.33 ≈ 0.064495%, CI: 0.15 ± 1.96 × 0.0644 ≈ [0.0237, 0.2763]
Same case as free speech, hard to interpret because EA History was not significant from the start despite the overlapping CIs.
So all we know for certain is that when it comes to political liberties, electoral authoritarian regimes like women's rights as much as democracies do, cringe!