Appunti esame Migration in Europe mod.1 prof. Venturini

OLS

Suppose we have a sample of N observations

on individual wages and personal

characteristics

Wage (salario), age and gender

How in this sample wages are related to other

observables?

OLS general equation:

In our empirical case: 15

Migration in Europe – Venturini a.a 2017/2018

B0 è l’intercetta con asse y, B1 ampiezza retta.

Puntino sopra retta positive error, sotto la retta negative error.

I puntini sono le varie osservazioni, la retta è la mediana.

16

Migration in Europe – Venturini a.a 2017/2018

Il margine di errore sono i puntini distanti. Più sono distanti più è ampio il margine di errore.

Minimize the error sia positive che negative, precision of the answer.

le informazioni in modo ennesimo e quindi infinito, e minimizza l’errore

La sommatoria è di tutte

in questo modo. 17

Migration in Europe – Venturini a.a 2017/2018 18

Migration in Europe – Venturini a.a 2017/2018

8 Marzo

How to measure migration: problems. Connecting questions to data choice.

Migration is a very complex phenomenon to measure and to follow.

There is

➢ departure,

➢ arrival,

➢ move to another country,

➢ move back,

➢ departure again,

➢ arrival again

But who are the migrants? International or Internal?

Measuring migration: key concepts

- Who do we measure? Migrants definition issue.

Geopolitical-legal criteria:

1. Country of usual residence (UN definition)

2. Country of birth: population born aboard

3. Country of citizenship: foreign population regardless its country of birth. Country of

citizenship at birth matters!!

Place of residence

Two main issues in defining and measuring the place of residence:

1. Duration of stay: very different, depend on administrative rules of the different countries. At

international level, the one-year duration is strongly recommended.

2. Legal status

Citizenship

The definition of acquisition of citizenship used by EUROSTAT includes changes from a previous

citizenship or no citizenship to that of the country of residence.

19

Migration in Europe – Venturini a.a 2017/2018

Some citizenship acquisition involves the loss of previous citizenship, others do not.

The definition encompasses changes that have been applied for by the individuals, but not i.e.

citizenship granted at birth.

Eligibility to acquire citizenship and the procedures involved vary between countries and may

change over time.

Clearly any such changes may result in discontinuity in the patterns of citizenship acquisition in the

country concerned.

Moroccan emigration stocks in France

2005

“Immigrés” people born aboard and

then go to France.

Why? Many countries lack a system for the continuous registration of international migrations (or

do not publish the data). Among the countries that process the data varies the scope and the

meaning.

- Technical reasons (difficult to distinguish between migrants and commuters etc.)

- Bureaucratic reason (they do not change residency)

- Political reasons

Stock of migrants in Spain

Difficult to cover this phenomenon. The

stock are much bigger (ex WB data).

Who do we measure?

Duration criteria (UN definition)

- Short-term migration: at least 3 months but less than 12

- Long-term migration: at least 12 months

BUT

- Comparability issues

- Include movements different in nature

Questions:

How many foreigners live in a country? 20

Migration in Europe – Venturini a.a 2017/2018

How many arrived?

How many have left?

What do we measure?

- Stock: n. of migrants living in a country at time T (e.g. 1 Jan. 2014)

- Flows: n. of migrants who move in a specified period (usually one year)

Flow net flows or gross flows?

- Inflow immigrants: out flows of migrants

- Net inflows: gross inflows gross outflows

- Population change (different between two stock values) is due natural change of the

population (fertility minus mortality) and net migration (inflows minus outflows)

Net inflows can be + o - : if positive = destination country; if negative = destination country.

The overtaking by migration

Green: natural change of population =

–

birth death

Blue: total population change

Red: net migration + statistical

adjustment

The point is ’93, until that day the

fertility was higher than migration.

How do we measure?

Destination vs origin statistics (direct vs indirect): counting who is in.

- Emigration from a given country should be reconstructed as the aggregated immigration

originating from his country in all other countries of the world

- Immigration in a given country should be measured using data collected in this country

Who counts?

-The Institution of destination countries

are

1. the census: control of the size of the

population.

2. Also residence permits =

administrative information (Spain you

can register at a local office also if you

are irregular).

3. And ad hoc surveys (used more).

- The Institution of origin countries are:

1. Consulates = voluntary registration.

2. Ad hoc surveys = mix of second generation migrants and first-generation migrants

21

Migration in Europe – Venturini a.a 2017/2018

➢ Traditional immigration

countries: more outflow (ex.

Australia, USA)

➢ Eu countries characterized by

large inflow due to post-war

labor recruitment

➢ Eu countries (ex. UK e FR e

Netherlands and Belgium) long

immigration histories linked to

their colonial past.

➢ Eu countries of “recent

immigration (ex. Italy, Spain)

Who is a migrant in the EU statistics?

Based on the change of usual residence.

It only implies:

- Crossing of (current) borders

- Duration of actual/intended stay of at least 1 year

Changes of usual residence between EU MSs are migrations. It does not include short-term

migrations (3-12 months).

Asylum seekers not included in population counts in many MSs (unlike refugees).

How many migrants in/to EU-28?

Emigration in 2013: more in MSs than EU-28. 2.8 vs 1.4.

In 2013 net migration + statistical adjustment: 1.7 in EU-28 and 1.7 in Sum MSs.

Stock on 1.1.2014 of foreign born persons is more in MSs than EU-28.

22

Migration in Europe – Venturini a.a 2017/2018

Migrants stocks on 1.1.2014

Samuel Nocito pt2

Dependent variable: wage

Error term: difference between the observer data e OLS line

Either dependent or independent variables can be:

The interpretation of the coefficient estimates changes according to the combination of these types

of variables.

Model A: continuous dependent variable 23

Migration in Europe – Venturini a.a 2017/2018

Model B: dummy dependent variable.

Model C: logarithm dependent variable.

This is a model of type A: continuous dependent variable.

B1 refers to a DUMMY explanatory

variable (Male). The movement of

Male from 0 to 1 produces a B1 unit

change in Wage.

There are four different types of data:

- Cross-sectional: sample of observations taken at a given point in time (ex. Counts)

- Time series: observations on a variable or several variables over time (ex. Follow people to

their life)

- Pooled cross-sectional: different random samples are asked the same questions over time

- Panel (or longitudinal): consists of a time series on a same individual

24

Migration in Europe – Venturini a.a 2017/2018

Endogeneity occurs whenever the explanatory variable (regressor) is correlated with the error term.

Endogeneity conditions:

- Measurement error: error made in measuring the dependent or the explanatory variable.

Ex: wages is an information that people not always want to provide. Difficult to measure the



sample information data itself correlated with the error.

25

Migration in Europe – Venturini a.a 2017/2018

Correlation is a statistical measure describing the size and the direction of a relationship between

two or more variables.

Causality indicates that one event is the result of the occurrence of the other event.1

Example 1: Smoking might be correlated with alcoholism but it is not a cause of it.

Example 2: Immigration might be correlated to the total level of crime in a specific region or

province, however it is not a direct cause of it (see next example).

⇒

Causality is compromised by endogeneity other driven factors affecting the choice.

26

Migration in Europe – Venturini a.a 2017/2018

9 Marzo

How many EU born migrants?

Third national migrants (red)

Born in another MS (blue)

Reason to migrate:

- Economic reasons

- Refugees

- Family reunification

- Students

- Tourists

Free mobility is the component of migration flows that is most sensitive to economic conditions

The large part is family member. Main gate of entrance.

Work is a minority. The labour workers are in minority in comparison to family member.

Across EU countries, there is a large dispersion in the composition of new migration

Sweden small country, a lot of humanitarian.

Germany small share, a lot of free movement (EU citizens).

UK work important.

Family reunification increase stock of

migrants.

A comparison of labour force survey

and of OCDE standardized permit data

(non-EU migrants) 2005-08

Country of new immigration: IT,

Portugal and Ireland: large family

reunification. Increasing because of

new family member of foreign

workers are arriving. 27

Migration in Europe – Venturini a.a 2017/2018

Reason for migration is the most important determinant of outcomes for new arrivals, but there is

some convergence over time

Family: 355 the more they stay the more they are able to find a job.

Reason for migration: the single most important determinant of outcomes

An important factor for policy: only labour migrants are directly selected. But category of migration

is rarely considered in analyses of migrants’ outcomes.

Non-EU migrants who come for work or study have higher qualification levels than family or

humanitarian migrants, but still lower than the native-born.

The returns to qualifications are higher for labour migrants (and for former international students)

than for the other two groups but still tend to be lower than for the native born.

Type of migration 1

- Permanent

- Temporary

- Seasonal

- Circular

- Return migrants

- Highly skilled

- Low skilled

Stock of migrants by origin country in main destination areas in 2000

Top immigration countries in 2010

Top USA, Russian Federation, Germany, Saudi Arabia, ..

Percentage of population

Top Qatar, Monaco, United Arab Emirates, Kuwait, ..

28

Migration in Europe – Venturini a.a 2017/2018

Main citizenships of EU and non-EU foreigners

residing in the EU-27 in 2010

One quarter of foreigners in the EU are citizens

of Turkey, Roma