We summarize the distribution of staff child ratios across markets with and without R&Rs, for each of the four age groups, in Table 3. These raw data suggest that markets with R&Rs have less dispersion, lower maximum and lower average staff/child ratios. As with prices, the effects may be explained by differences in consumer demographics or input costs that are correlated with the presence of R&Rs. Recall that markets with R&Rs tend to have higher incomes, are more educated, have higher rents and are more urban than markets without R&Rs.
To control for other effects on the distribution of staff/child ratios, we estimate the reduced form models described in Section 4. Methodologically, we proceed in a manner analogous to the one adopted for the study of price distributions. For each dependent variable (coefficient of variation of stafi/child ratios, maximum staff/child ratio and average stafi/child ratio), we estimate separate models for infant, toddler, preschooler, and school-age groups, using ordinaiy least squares with market-level data. Standard errors are heteroskedasticity robust. We also estimate the effect of R&Rs on average staff/child ratios using center level data. For brevity, we summarize the main results from these fifteen regressions in Table 7, which includes the estimated coefficients with associated t-statistics on RANDR.
Our findings indicate that R&Rs have no significant effect on stafi7child ratios. Moreover the point estimates are fairly small in absolute value, though they are, for the most part negative. This is an important finding since it suggests that the beneficial effects of R&Rs on price distributions are not offset by a significant deterioration in observable quality. We cannot, of course, say anything about effects on unobservable quality. payday loan companies