Hello, dear friend, you can consult us at any time if you have any questions, add WeChat: daixieit

ECO00003H

Applied Economics

2020-1

SECTION A

1. What are the economic benefits of targeting current government policy towards reducing youth unemployment?  In the middle of an economic crisis which policies might prove most effective at reducing youth unemployment?

An important part of this essay is a discussion of the empirical evidence that an experience of youth unemployment causally lowers outcomes for the individual later in life.

If unemployment carries a persistent scar then a policy aimed at youth will raise their future earnings, lower benefit dependency, increase taxation revenues.  

There is a negative correlation between the incidence of youth unemployment and later labour market outcomes.  But is this a causal effect?  

Good essays will discuss intuition for the endogeneity problem, explaining that unobserved traits, eg u (ability/tastes etc.) drive youth unemployment and the later wage. Explaining using equations is important.

Various papers have attempted to identify the causal effect and the essay should discuss the methodology of at least one paper with a critique of the method.

On top of this, the students can decide which other arguments to bring into their essays, for example

- Citing the statistics on youth unemployment during a recession

- Macro consequences of youth unemployment

- Theories predicting a scar from youth unemployment

Finally, there are different policies which have been shown to reduce youth unemployment. The students can go into detail if they choose, or quickly discuss if they have focused most of their essay on the empirical evidence.

2. Does economic evidence suggest that the correlation between the success of parents and children is driven by nature or nurture? Given this evidence, what are the implications for inequalities observed in society across different ethnic groups?

The model

 

Where y denotes earnings, c denotes child and p denotes parent.

If there is a link between earnings of parents and children, the potential mechanisms are:

1) Genetics: nature

2) Nurture. For example, high income buys: a house in an area with good schools, extra tuition, nepotism which can all help the child to earn higher income themselves.

To estimate the causal or policy effect, evidence needs to eliminate the influence of genetics.

Discuss critically whether the causal evidence suggests that there is a strong policy role.  

a. Adoptee evidence

b. Oreopolous et al reply on firm closures

I want students to critically at least one paper which requires explaining the methodology and the weaknesses.  The evidence does not all give the same answer so students are required to reason which study is more plausible in order to answer the essay question.

Next, different ethnic groups have different intergenerational mobility: this evidence was discussed in the lecture notes considering Chetty et al (2019): larger intergenerational gaps for black men. Link this evidence to the question.

Overall a strong essay will critique papers and use the evidence to answer the specific question.

3. Economic evidence on the effect of immigration on native wages and employment tends to find very small or zero effects. Explain using a simple economic model why this might be the case and critically assess the evidence.

The simple model by Dustmann et al should be discussed clearly which shows that the aggregate effect of migration on native wages is zero; although there are potentially redistributive effects.

Evidence is needed to test the theory.  Students must define the causal effect and explain clearly the endogeneity problem.

One common method to identify the causal effect is instrumental variables.

Discuss the methodology and weaknesses of a set of papers in order to draw a conclusion.

A very strong essay will answer the specific question using economic arguments and evidence.  Critique of the empirical methods is important. 


SECTION B

4. The new Biden administration in the United States is proposing to raise the national minimum wage to $15/hour. What does your knowledge of the economic theory and evidence tell you about the likely effects that such an increase will have on the level of employment in the United States?

A similar proposal was made by the Obama administration and is the subject of the paper by Aaronson, D. and E. French. How Does a Federal Minimum Wage Hike Affect Aggregate Household Spending? (2013), Chicago Fed Letter, No. 313, discussed in the lectures. Looks at the overall impact on the US economy of an increase in the (federal) national minimum wage from $7.25 to $9, an increase of $1.75.

(a) Assess the number of workers whose wages would be affected

(b) Predict the likely effects of an increase in the hourly federal minimum wage on total household income, consumer prices, and aggregate household spending.

Shows that a $1.75 increase in the minimum wage could raise real GDP by about 0.3 percentage points in the short run (one year).

They are skeptical about the direct impact of the increase in the minimum wage on employment which they assume is negative. The impact of an increase in the federal minimum wage on GDP and employment overall would be more positive still if the Card and Kruger results hold. In the lectures the positive Card and Kruger estimates of the impact of the increase in the minimum wage on local employment are discussed as an example of the result of monopsony amongst employers:

 

Where this model is tested in the burger industry in New Jersey:

On April 1, 1992, New Jersey raised the state minimum wage from $4.25 to $5.05 an hour. Card and Krueger (C&K) collected data on employment at fast food restaurants (minimum wage employers) in New Jersey in February 1992, before the change and again in November 1992, after the change. They collected the same data in eastern Pennsylvania, a bordering state, where the minimum wage stayed at $4.25. C&K applied the Difference-in-Differences (DiD) strategy to estimate the effect of the minimum wage increase on employment levels.

 

So if the C&K results are correct then the impact of the Biden increase in the federal minimum wage should be greater.

Students should provide a thoughtful and critical response which is well supported by the empirical evidence discussed in the lectures. See the department marking guide with respect to more detailed and generic marking guidance.

5. In the next few years we can expect that there will be (A) an increase in the numbers of police officers and (B) an increase in unemployment in the UK. Using your knowledge of empirical evidence and the economic model of crime, what do you think the effects of these two developments will be on the level of crime in the UK?

Compare and contrast Supply of and Demand for crime. (A) applies to demand and (B) applies to supply.

Demand for crime can be reduced by:

Prevention activities (locks on your doors, police numbers)

Threat of incarceration

Supply of crime can be reduced by:

Assessment of the benefits of crime

gain from committing the crime

commit crime if benefit exceeds wage rate

unemployment a determinant

For (A), questions raised at the introduction to this topic in lectures include: Does an increase in police force reduce crime rates? Higher police numbers reduces the "demand" for crime; "deterrence technology”. What can we tell from comparing police numbers to crime rates?

Very hard to prove anything causal because of the endogeneity of police and crime: e.g. more police deployed because crime rates are high. Even exploiting a change in police numbers isn’t good enough – could have correlated shocks in (un)observable crime and police. For identification, to show causality, we need to exploit an unexpected change in police numbers. Students should be aware of these issues and take a thoughtful approach to the questions.

The response to the 7-7 bombings in London led to 30% increase in police deployment over next 6 weeks. But only in particular London boroughs. It was an unexpected change so we can compare "before“ data to "post" data in affected areas vs other areas. Sudden "switch off" of policy at 6 weeks.

Want them to think carefully about the identification assumptions here. Authors have to prove

1. Police deployment increased in targeted areas

2. Terrorist attack didn’t directly drive crime rates in targeted areas

3. Terrorist attack didn’t cause people in targeted areas to behave differently (stay inside, avoid public transport etc.)

Used this paper in lectures: Draca, M. Machin, S. and Witt, R. (2011) “Panic on the Streets of London: Police, Crime and the July 2005 Terror Attacks”, American Economic Review 101: 2157-81. They found a fall in crime rate of 11% in treated group (policy on - policy off) compared to control group in the 6-week period of extra policing. Minimal evidence of persistence in the period afterwards. They found a significant effect of the policy in targeted areas in the 6 weeks period of extra policing for Thefts and Violence; not for Sex crimes, Robbery, Burglary or Criminal Damage. This is contrary to many other smaller scale papers which find no effect - but changes in police numbers were often anticipated in other studies.

So the evidence is in favour of an impact of increased police numbers on reducing the demand for crime.

For (B) Bell, Blinder and Machin look at the effect of the local unemployment rate when individuals enter the labour market (age 16) on criminal activities. This study aims to identify the causal effect of "opportunities" on patterns of crime.

Hypothesize that early labour market conditions are important determinants of youth crime.

At higher unemployment rates, the expected returns to work fall. Ceteris paribus this should increase youth crime rates.

BUT: also expect an effect of age 16 unemployment rates on later crime rates

Early exposure to crime increases criminal expertise, may lower wage rate (through criminal record)

 

 

Therefore an increase in unemployment can be expected to increase supply of crime which can be expected to be persistent. So (A) and (B) provide offsetting effects; say something about whether they expect the effects to balance out and how you would judge this – eg the short and longer term effects.

Students should provide a thoughtful and critical response which is well supported by the empirical evidence discussed in the lectures. See the department marking guide with respect to more detailed and generic marking guidance.

6. Richard Layard (LSE) has recently written that “Population surveys have demonstrated that significant increases in average income in many countries have not translated into a corresponding rise in average levels of happiness and subjective wellbeing.” Explain how this paradox can be reconciled with the findings of studies of individuals which find that personal income and satisfaction are positively related.

This question is essentially about the Easterlin Paradox which needs to be explained in any reasonable answer. In the lecture notes, I said:

In many countries average happiness has remained constant over time despite large rises in GDP.  For example, between 1973 and 2003 happiness measures remained fairly flat in the US whilst real income per capita almost doubled.

 

Does the Easterlin Paradox mean that additional income buys little if any extra happiness? Is it the case that once an individual rises above a poverty line or ‘subsistence level’, the main source of increased well-being is not income but, for example, friends and a good family life? This ‘subsistence level’ has been argued to be as low as US$10,000 per annum. Radical implication for developed countries is that economic growth per se is of little importance, and should therefore not be the primary goal of economic policy (Oswald, 1997). Layard (2005) goes as far as arguing that we need a ‘revolution’ in academia, where every social scientist should be attempting to understand the determinants of happiness, and it should be happiness which is the explicit aim of government intervention rather than GDP. Yet, at a given point of time within a country, individual income correlates positively with measures of happiness and subjective well-being.

But stylised facts from the happiness literature using individual data suggest a positive relationship between own income and life satisfaction. I want the students to show clear and critical understanding of these ideas.

If individuals care more about their relative income (their income relative to the income of a reference group) than about their absolute income, then this can explain the Easterlin Paradox. Relative income differs within a cross-section in a given point of time, but if a society as a whole gets richer, then relative income does not necessarily change. Research on happiness and subjective well-being therefore pays particular attention to relative income and reference groups (the group/s with which people compare themselves).

 

Want them to talk about reference groups: social comparison/status (reference group defined by peers); and adaptation/habituation (reference group defined by oneself in the past).

 

We focused on this paper in the lectures: Card, D. Moretti, A. and Saez, E. (2012).  Inequality at work: The effect of peer salaries on job satisfaction” American Economic Review 102(6), 2981-3003. I would expect them to talk about these findings to support their work, noting their efforts to show causality. A really good student might also refer to Bond, T. and Lang, K. (2019). The sad truth about happiness scales. Journal of Political Economy 127(4), 1629-1640. And discuss difficulties using Likert scales in this type of analysis. See the department marking guide with respect to more detailed and generic marking guidance.