Intermediate Macroeconomics Questions, due on April 29, 3:00pm (UTC+8)

1. Redistributive government policies.

a) Overview

In the introductory lecture, we have presented evidence that government policies have a major offsetting effect against rising income inequality. Figure 1 captures this offsetting effect by the difference between the Gini coefficient of market income and the Gini coefficient after taxes and transfers.

Figure 1

Source: IRPP (2016)

In this lecture, we will explore the role of redistributive government policies as a mitigating factor against rising market income inequality. Redistributive policies are ‘tax and transfer’ policies, in which use tax revenue to fund social program (e.g., childcare) or to provide income support to individuals (e.g., social assistance program, employment insurance).

Redistributive government policies commonly use as tools:

· Taxing income of high-income earners,

· Provide transfers (or income subsidies) to low-income earners.

Note that other types of government policies, such as regulatory policies, could also have a redistributive impact. However, regulatory policies they are not redistributive, i.e., tax and transfer, by design. For instance, the immigration policies we have touched on are regulatory in nature but could have major redistributive impact on the capital and labour income shares.

b) Types of redistributive programs

Figure 2

Source: IRPP (2016)

Figure 2 represents five major redistributive programs run by federal and provincial governments.

In this lecture, we will analyze the impact social assistance/employment insurance and child benefits programs on labour outcomes. In a subsequent lecture, we will discuss the economic impact of government pension plans (CPP/QPP) and old age security.

The use of policies that encourage (e.g., subsidies, benefits) or discourage (e.g., taxes) a particular type of behaviour that may unintentionally impact employment even if their objective is to redistribute income.

Figure 2 panel b) reveals that the spending on child benefits programs is relatively stable over time with a structural break in the earl 2000s when Quebec introduced the government-run childcare program. In contrast, other provinces continued with the existing policy of providing child benefits.

This policy change in Quebec serves as a natural experiment, in which Quebec serves as a treatment group, while all other provinces serve as a control group. Analyzing government policies in a natural experiment allows us to credibly elicit the true impact on parents’ employment of introducing childcare relative to receiving child benefits.

Figure 2 panel b) also reveals that the employment insurance and social assistance are subject to large variation that could be decomposed into:

· a structural break in the early 1990s when governments cut down on the generosity of employment insurance but increased the generosity of social assistance,

· cyclical fluctuations that overlap with rising involuntary unemployment during recessions.

The generosity of a social program refers to the size of the transfer payments.

c) Incentives and opportunity cost

In this section, we will develop the concept of an opportunity cost to explain why individuals’ choice whether to accept employment could be influenced by a redistributive government policy.

To evaluate the decision of a citizen whether to be employed or rely on social assistance, we will develop the concept of an opportunity cost.

Consider two options: receive social assistance or be employed.

The opportunity cost of receiving social assistance measures the net foregone amount of the highest-valued alternative.

In the presence of two options, 1) receiving social assistance and 2) being employed, the highest-valued alternative to receiving social assistance is being employed.

We will explore two closely related numerical examples, in each of which we will assume that being employed is the socially efficient outcome.

Employment $ 2,500 per month,

Employment Insurance $ 2,000 per month,

Childcare cost if working $ 1,600 per month.

Our task is to calculate the opportunity cost of receiving social assistance based on the provided information.

Opportunity cost of receiving social assistance = Net gain of being employed – Net gain of receiving social assistance

Net gain of being employed = 2,500 – 1,600 = 900

Net gain of receiving social assistance = 2,000

Opportunity cost of receiving social assistance = 900 – 2,000 = -1,100 < 0.

Since the opportunity cost of receiving social assistance is negative, it is in one’s self-interest to receive social assistance in the form of employment insurance.

Numerical example:

You are provided with the following information.

Employment $ 2,500 per month,

Employment Insurance $ 2,000 per month,

Childcare cost if working $ 1,600 per month,

Childcare subsidy $ 1,500 per month,

Our task is to calculate the opportunity cost of receiving social assistance based on the provided information.

Net gain of being employed = 2,500 – 1,600 + 1,500 = 2,400

Net gain of receiving social assistance = 2,000

Opportunity cost of receiving social assistance = 2,400 – 2,000 = 400 > 0.

Since the opportunity cost of receiving social assistance is positive, it is in one self-interest to be employed.

2. Natural experiments.

a) Overview

We will explore the economic impact on employment of two redistributive policies implemented by the Québec government.

In the first natural experiment, Lemieux and Milligan (2008) explore whether providing a more generous social assistance adversely impacts employment.

In the second natural experiment, Lefebvre et al (2009) explores whether subsidized childcare positively affects parental employment.

b) Social assistance

· Nature of natural experiment

Lemieux and Milligan (2008) exploit the difference in the generosity of social assistance across different age groups to study its impact on employment.

Figure 3 demonstrates that until 1989 the benefits differed substantially between the childless social assistance recipients in Québec under age 30 and those over age 30.

Figure 3

Source: Lemieux and Milligan (2008)

In general, the average labour characteristics of the two age groups are vastly different with respect to educational attainment and work experience. However, a comparison of the two age groups in proximity to their 30th birthday allows to compare the employment outcomes of individuals with virtually identical average labour characteristics but offered different amounts of social assistance benefits.

· Results

Figure 4 presents evidence that employment drops abruptly once individuals become eligible for the higher social assistance benefits at the age 30.

This result suggests that the generosity of social payments has an adverse impact on the employment rate, which abruptly drops from 66 % to 61 % at age 30.

This provides evidence that policies that offer benefits (or costs) affect economic behaviour by influencing one’s choice whether to be employed or to rely on government assistance.

Figure 4

Source: Lemieux and Milligan (2008)

c) Subsidized childcare policy

· Nature of natural experiment

The government of Québec introduced in 1997 a subsidized childcare policy ($5 per day per child) that:

· Initially targeted the 4-year olds,

· was gradually expanded to younger cohorts in subsequent years.

The subsidy covered approximately 75 – 80 % of the cost for childcare services (Lefebvre et al, 2009).

For simplicity, we will focus our attention to the initial phase of the childcare policy targeting the 4-year olds.

The objective of this study is to:

· report any increase in the utilization of childcare spots both in the short-run and in the medium-run relative to the period prior to the policy implementation,

· link any increase in the utilization of childcare spots to corresponding increases in employment among mothers both in the short-run and in the medium-run.

Figure 5

Source: Lefebvre et al (2009)

The study uses the other Canadian provinces as a control group since no changes in childcare policy were implemented in the rest of Canada during the examined period.

· Results

· Direct effect

Figure 5 indicates that there is a sizable increase in the mean hours per week for the children eligible for subsidized childcare both in the short-run and in the medium-run.

This result indicates that the subsidy changes parents’ behaviour by incentivizing them to utilize childcare services.

· Spillover effect: Employment

Figure 5

Source: Lefebvre et al (2009)

There is a sizable increase in the number of weeks employed for the mothers with children in the eligible age group both in the short-run and in the medium-run.

The economic significance of this effect is substantial. It increased the labour supply of women by 3.8 % (approx. 70,000 mothers), which translates to 1.7 % increase in GDP per capita.

There is also a persistent effect on mothers’ employment rate, which increases not only during the child’s early years (attending childcare) but also in the child’s later years (attending elementary school).

This effect is more pronounced for the mothers who have completed high school education or less. Higher educational attainment is highly correlated with wage, which suggests that mothers employed in low-wage occupations, were mostly likely to rely on government assistance.

This persistent effect could be interpreted that subsidized childcare prevents raising a child to come at the cost of a parent leaving the labour force for an extended period beyond pregnancy and birth.

· Spillover effect: Government revenue

The most surprising effect of subsidized childcare is that the program generated more government revenue than it cost to subsidize childcare. The two main sources of government revenue are:

· Increased income tax revenue from mothers’ joining the labour force,

· Decreased spending on social benefits due to mothers’ joining the labour force.

Ultimately, this surprising effect on government budgets is due to the persistent effect of mothers participating in the labour force that they would have not done in the absence of subsidized childcare.

Table 1 presents the impact on government revenue, where the static effect refers to the time period when a child is enrolled in childcare and the dynamic effect the time period after a child is enrolled in childcare.

Table 1

Source: Fortin et al (2012)

d) Key insights

Redistributive policies must be carefully designed by noting their incentive effects on employment.

A childcare program contributes to continuous employment of mothers and, in turn, reducing the gender income gap between males and females. A heavily subsidized childcare program alters the incentives in favour of being employed.

In contrast, the impact of social assistance is a double-edged sword. For the involuntarily unemployed, more generous social assistance payments contribute to reducing income inequality.

However, the program also generates an incentive effect by encouraging some to claim unemployment benefits instead of working. This incentive effect, in fact, contributes to rising income inequality as social assistance recipients receive a smaller income than if employed.

2