Narrow Results

This paper analyzed whether the child quantity-quality (CQQ) trade-off is applicable in the case of Malaysia. Utilizing the instrumental variable (IV) method, our analysis produces results that are consistent with the hypothesis that the trade-off is unlikely to be applicable to Malaysia as a whole due to the generous public provision of education. However, the results show that the CQQ trade-off exists for the high-income group. Taken together, if both a larger stock of human capital and population are desired, the government should continue to provide education for its people while at the same time maintain policies that stimulate growth.

Using unique survey data on rural secondary school children, this paper evaluates the relative quality of Islamic secondary schools (i.e., madrasahs) in Bangladesh. Students attending registered madrasahs fare worse in maths and English than students attending non-madrasah schools. However, failure to account for non-random sorting overestimates the negative influence of madrasahs on student’s achievement. Evidence on the magnitude of this bias is presented. Once selection effect is taken into account, the madrasah’s disadvantage in English is small while that in maths becomes insignificant. Given the overall low level of achievement, this suggests that madrasah students perform just as poorly as those from non-madrasah schools in rural Bangladesh.

This study examines R&D and information and communication technologies (ICTs) in the innovation process, and how these two are integrated with each other for innovation by panel data analysis. The surveys were conducted in February 2012 and March 2017. The number of innovations is taken as an outcome variable, while explanatory variables related to R&D and ICTs were extracted from related questions by factor analysis. The fixed effect robust model with an instrumental variable is estimated, since the error term may contain heteroscedasticity. R&D autonomy and the cross-term of R&D autonomy and ICTs are significant, indicating that ICTs contribute to innovation via R&D autonomy.

Municipal water demand has declined over the past several decades in many large cities in the western United States. The same is true in Clovis, New Mexico, which is a small town in arid eastern New Mexico, whose sole water source is from the dwindling southern Ogallala Aquifer. Using premises-level monthly panel data from 2006 to 2015 combined with climate data and additional controls, we apply a fixed effects instrumental variable approach to estimate municipal water demand. Results indicate that utility-controlled actions such as price increases and rebates for xeriscaping and water saving technology have contributed to the decline. Overall water demand was found to be price inelastic and in the neighborhood of $-$0.50; however, premises receiving toilet and washing machine rebates were relatively more price inelastic and premises receiving landscaping rebates were more price elastic, though still inelastic. In addition, the average premises receiving its first toilet rebate reduced water use by 8.4%, washing machine rebates lowered use by 9.2%, and the average landscaping rebate reduced water use by less than 5.0%. From the utility’s perspective, and assuming a 5.0% discount rate, levelized cost analysis indicates that toilet rebates are 34% more cost effective than washing machine rebates and nearly 800% more cost effective than landscaping rebates over their respective lives per volume of water conserved. While this research focuses on Clovis, estimation results can be leveraged by other small to mid-sized cities experiencing declining supplies, confronting climate change, and with little opportunity for near-term supply enhancement.

I find that downward bias of the estimated coefficient of betas in the Fama-MacBeth cross-sectional regression is caused by endogeneity of the estimated betas due to measurement errors. I propose an instrumental variable methodology that purges the endogeneity. The purged betas have a 95% correlation with the original betas and retain the relation with firm size. I document that controlling for the purged betas in the Fama-MacBeth cross-sectional regression has higher statistical power to correctly reject the null hypothesis of nonexistence of size premium and also has higher R-squared and higher total sum of squares than the Brennan-Chordia-Subrahmanyam (1998) methodology.