What Providers Changed—and What They Did Not
The Illinois CACFP study asks a specific question about home daycare providers: after some moved from the lowest to the highest reimbursement schedule, which parts of their offered operation changed? The answer is not a single measure of “more childcare.” It is a pattern across dinner, shift timing, age-specific variety, prices, quality proxies, and local context. Reading those margins together shows how a meal subsidy can reach the clock while stopping short of a guarantee about cheaper care, better care, or care that families actually use.
Dinner moves first
The clearest food response is dinner. The estimated move to high reimbursement raises the probability that a provider offers dinner by 5.0 percentage points. The baseline mean is 0.173, so the estimate represents a relative increase of 28.9% over an already modest starting share. In practical terms, the treatment does not make dinner universal; it makes dinner more common among the providers in the comparison.
The selectivity of the response matters. The study finds no statistically significant change in the likelihood of offering breakfast, lunch, or a morning snack. The afternoon-snack estimate is a small decline of 1.6 percentage points and does not meaningfully offset the dinner increase. If higher reimbursement simply encouraged providers to add every reimbursable food item, we would expect a broad movement across the meal categories. Instead, the strongest response appears where the meal is most closely connected to late-day care.
A non-significant estimate should be read carefully. It means the study does not detect a change large or precise enough to distinguish from zero under its model; it does not prove that breakfast or lunch could never respond to a different policy or in a different setting. The result is still informative because the contrast among meal types fits the mechanism developed earlier. Daytime meals were already part of many providers' ordinary routines. Dinner had more room to be an entry point into a later service.
The measure is an offering indicator, not a plate count. A provider who marks dinner as offered may have made a new service available to some children, but the data do not establish how many children attended at dinner or how many meals were served. Nor do they measure the nutritional quality of the meals actually eaten. The strongest defensible statement is therefore that higher reimbursement increased the reported likelihood of offering dinner among the active providers studied.
The clock changes at the edge of the day
The schedule results line up with the dinner response. Providers receiving the higher reimbursement are 5.5 percentage points more likely to offer an evening shift. Because the baseline incidence is 0.098, that is a relative increase of 56.1%. The large relative percentage does not mean that most providers began evening care; it reflects a sizeable change from a low starting point.
The number of shifts offered rises by 0.064 from a mean of 1.139, a relative increase of 5.6%. That estimate captures variety rather than simply whether a provider has any shift at all. The indicators for day and overnight shifts do not change significantly. Taken together, the results point to a specific adjustment: the reimbursement is associated with more evening flexibility and a modestly larger menu of shifts, not with a general expansion in every direction.
The timing within the day is equally revealing. Average weekday opening time does not move significantly, while the reported closing point shifts later.