FSAR | Lab #7 | Fall 2025
28th November, 2025
Exploring the Effects of the Minimum Wage Law
1 Contrary to what economic theory predicts, we found no evidence of a negative impact of the minimum wage law on employment in New Jersey. Use the available data to investigate this issue further.
1.1 Did fastfood stores change their hours of operation in response to the increase in minimum wage?
1.2 Did fastfood stores raise meal prices in response to the increase in minimum wage?
1.3 Did fastfood stores change the fraction of full-time employees in response to the increase in minimum wage?
2 Identify three limitations in the data used and/or analysis we have conducted.
Get started by loading libraries, reading and preprocessing data.
library(dplyr)
library(ggplot2)
library(stargazer)
library(lmtest)
library(sandwich)
# Load data
load("data/fastfood.RData")
# Create indicator for missing data by ID
tb.fastfood <- tb.fastfood %>%
mutate(missing = as.double(complete.cases(tb.fastfood)==FALSE)) %>%
group_by(id) %>%
mutate(missing = max(missing)) %>%
ungroup()
# Keep only IDs with complete information
tb.fastfood <- tb.fastfood %>% filter(missing==0)
tb.fastfood <- tb.fastfood %>% select(-missing)
# Create state and time factors
tb.fastfood <- tb.fastfood %>%
mutate( state = factor(state, levels = c(0,1), labels = c("PA", "NJ"))
, time = factor(time, levels = c(0,1), labels = c("Before", "After")))
# Create a table with the variable means for treatment and control groups before and after the policy
tb.bf.aft <- tb.fastfood %>%
group_by(state, time) %>%
summarise(
mean_emptot = mean(emptot),
mean_wage = mean(wage),
mean_pmeal = mean(pmeal),
mean_hrsopen = mean(hrsopen),
mean_fracft = mean(fracft)) %>%
ungroup()
1 Contrary to what economic theory predicts, we found no evidence of a negative impact of the minimum wage law on employment in New Jersey. Use the available data to investigate this issue further.
1.1 Did fastfood stores change their hours of operation in response to the increase in minimum wage?
# Boxplot: Hours of operation before and after the law
ggplot(tb.fastfood, aes(x = time, y = hrsopen, fill = state)) +
geom_boxplot() +
labs(x = "Period", y = "Hours of Operation", fill = "State") +
theme_bw()
# Diff-in-Diff plot (mean hours)
ggplot(tb.bf.aft, aes(x = time, y = mean_hrsopen, group = state, color = state)) +
geom_point() + geom_line() +
labs(y = "Mean Hours of Operation") +
theme_bw()
# Difference-in-Differences regression
lm_hrs <- lm(hrsopen ~ state + time + state:time, data = tb.fastfood)
coeftest(lm_hrs, vcov = vcovHC(lm_hrs, type = "HC3"))##
## t test of coefficients:
##
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 14.49180 0.39502 36.6863 <2e-16 ***
## stateNJ -0.27220 0.42807 -0.6359 0.5251
## timeAfter 0.16393 0.55139 0.2973 0.7663
## stateNJ:timeAfter -0.14727 0.59756 -0.2464 0.8054
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Interpretation:
No statistically significant change in hours of operation attributable to the minimum wage increase.
1.2 Did fastfood stores raise meal prices in response to the increase in minimum wage?
# Boxplot: Meal prices before and after
ggplot(tb.fastfood, aes(x = factor(time), y = pmeal, fill = factor(state))) +
geom_boxplot() +
labs(x = "Period", y = "Meal Price", fill = "State") +
theme_bw()
# Diff-in-Diff plot (mean price)
ggplot(tb.bf.aft, aes(x = time, y = mean_pmeal, group = state, color = state)) +
geom_point() + geom_line() +
labs(y = "Mean Meal Price") +
theme_bw()
# Difference-in-Differences regression
lm_price <- lm(pmeal ~ state + time + state:time, data = tb.fastfood)
coeftest(lm_price, vcov = vcovHC(lm_price, type = "HC3"))##
## t test of coefficients:
##
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.071311 0.078155 39.2977 < 2.2e-16 ***
## stateNJ 0.316610 0.088706 3.5692 0.0003854 ***
## timeAfter -0.035574 0.105060 -0.3386 0.7350217
## stateNJ:timeAfter 0.113652 0.120369 0.9442 0.3454309
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Interpretation:
Evidence suggests that meal prices increased in New Jersey following the minimum wage hike.
1.3 Did fastfood stores change the fraction of full-time employees in response to the increase in minimum wage?
# Boxplot: Fraction of full-time employees
ggplot(tb.fastfood, aes(x = factor(time), y = fracft, fill = factor(state))) +
geom_boxplot() +
labs(x = "Period", y = "Fraction Full-Time", fill = "State") +
theme_bw()
# Diff-in-Diff plot (mean fraction full-time)
ggplot(tb.bf.aft, aes(x = time, y = mean_fracft, group = state, color = state)) +
geom_point() + geom_line() +
labs(y = "Mean Fraction Full-Time") +
theme_bw()
# Difference-in-Differences regression
lm_fracft <- lm(fracft ~ state + time + state:time, data = tb.fastfood)
coeftest(lm_fracft, vcov = vcovHC(lm_fracft, type = "HC3"))##
## t test of coefficients:
##
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.359005 0.031127 11.5336 <2e-16 ***
## stateNJ -0.024160 0.034389 -0.7025 0.4826
## timeAfter -0.033099 0.044590 -0.7423 0.4582
## stateNJ:timeAfter 0.050733 0.049372 1.0276 0.3045
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Interpretation:
There was no statistically significant change in the composition of the workforce.
2 Identify three limitations in the data used and/or analysis we have conducted.
The use of a telephone survey introduces measurement error and recall bias. More reliable administrative data would strengthen the analysis.
The short panel (only two time periods) limits our ability to verify the parallel trends assumption critical to difference-in-differences validity.
We do not account for entry/exit (openings and closings) of fast food stores, which could represent another important margin of adjustment to higher wages.