This study examines the impact of fiscal decentralization on social delivery policy response in Nigeria, utilizing panel data for fourteen states in Nigeria for the period covering from 2000 to 2019. Specifically, this study evaluates the impact of fiscal autonomy on social indicators such as infant mortality rate, maternal mortality rate and adult literacy rate. The panel vector error correction mechanism (PVECM) the granger causality test was employed as the estimation techniques. The result of the granger causality test to determine the direction of causality relationship among the variables in the estimated model showed that there was one way directional causality running from maternal mortality rate (MMR) to fiscal autonomy (FISA); adult literacy rate (ADLIT) to population (LPOP); population (LPOP) to maternal mortality rate (MMR), adult literacy rate (ADLIT) to maternal mortality rate (MMR), and education expenditure (LEDU) to fiscal autonomy (FISA). This means that maternal mortality granger caused fiscal autonomy in Nigeria. The result also implies that adult literacy rate granger caused growth in population, and growth in population granger caused maternal mortality rate in Nigeria. The error correction variable in the infant mortality rate equation has the coefficient of 0.069. This indicates that approximately 6.9 percent of the distortion in the system would be corrected each whenever the system moves away from equilibrium. This depicts a slower speed of adjustment mechanism from the disequilibrium in the short run to equilibrium in the long run. The estimated infant mortality rate equation has a very good fit on the data and very high explanation power, given the adjusted R-squared of 0.736. The adjusted R-squared of 0.736 showed that approximately 74 percent of variation in the dependent variable (infant mortality rate) was accounted for by variations in the independent variables. This result is not in agreement with the theoretical postulate. In real term, the result showed that an increase in education expenditure by one percent would lead to a decrease in adult literacy rate by 0.22 percent, other things being equal. This result showed that spending in education has not improved the literacy rate in Nigeria. This suggests that it is either the fact that spending in education by the government has been so small that it cannot bring about improvement in the literacy or that funds meant for educational projects are siphoned into private pause, thereby resulting to huge decline in terms of literacy rate. Lastly, growth in population has a declining effect on adult literacy rate in Nigeria. This inference is in accordance with a priori expectation showing that as population increase, pressure is being put on the existing educational facilities and overwhelm it, leading to the decline in adult literacy rate. In concrete term, an increase in population by one percent resulted to a decrease in adult literacy rate by 0.73 percent, other variables remaining the same.  Based on our result, the study made some policy recommendation for the states to double their efforts in generating their internal revenue to become fiscally autonomous less reliance on the federal government for allocation. Also, there is need for the government to raise her expenditure in education and health sectors so as to increase the literacy rate and health sector outcomes in the country.