{[ promptMessage ]}

Bookmark it

{[ promptMessage ]}

Methods in Labor Economics

Lee 2004 underscorethe importance of avoiding using

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: and David S. Lee (2004) underscorethe importance of avoiding using conventional standarderrors in a discrete RD design, such as this one, which depends on a specific functional form to compute the cohort trend. predictedbreakin educationattainmentof 0.44, correspondingto Figure 2A. The t-statistic for rejecting the discontinuityis 6.5. Figure 5 shows the analogous graph for NorthernIreland,where the school-leaving age increased from 14 to 15 in 1957. The plotted averages are somewhat less smooth than for Britain because of the smaller sample size (R2 = 0.989). The quartic polynomial fit, also presented in Table 1, predicts an increase in average education attainmentof 0.397 years in 1957, indicatedby the vertical line. Figures 6 and 7 plot the correspondingraw mean log earnings for the British and Northern Irish samples for 32- to 64-year-olds who were 14 years old between 1935 and 1965. Earnings are measured in 1998 U.K. pounds using the Retail Price Index.'13 arnings from the comE 13 Note that the cross-section panel of the 1983 to 1998 panel includes fewer older cohorts at younger ages. For example, the GHS observes all those older than 51 who were 14-year-olds in 1945, but only those 14-year-oldsof 1945 who are older than 61. This discrepancy does not affect testing for a trend break in mean earnings following the introductionof the more restrictiveschool-leaving age, but it does lead to an upwardtrendin earningsfor younger cohorts. To account for the sensitivity of the results to the different age composition by birth cohorts, I also estimate the discontinuitywith age polynomialcontrolsand age fixed effects. Age fixed effects are possible with cohort effects VOL.96 NO. I OREOPOULOS: ATE COMPULSORY ESTIMATING AND LATEWITH SCHOOLING LAWS 163 LocalAverages Parametric and Fit 9.4 Pounds) UK 9.2 (1998 9 Earnings 8.8 Annual of Log 8.6 1935 1940 1945 1950 YearAged 14 LocalAverage 1955 1960 1965 PolynomialFit FIGURE 6. AVERAGE ANNUAL LOG EARNINGS BY YEAR AGED 14 (Great Britain) Note: Local averagesare plotted for British-bornadults aged 32 to 64 from the 1983 to 1998 GeneralHousehold Surveys. T...
View Full Document

{[ snackBarMessage ]}