Topic 4.1: Unemployment Patterns PDF Fall 2024
Document Details
![AppreciatedUranium](https://quizgecko.com/images/avatars/avatar-1.webp)
Uploaded by AppreciatedUranium
University of Bern
2024
Jeanne Tschopp
Tags
Summary
This document is lecture notes on unemployment patterns in labor economics, from Jeanne Tschopp at the University of Bern Fall 2024. The lecture notes overview unemployment and discuss job flows and worker flows to understand observed unemployment patterns within and across countries.
Full Transcript
Topic 4.1: Unemployment Patterns Labour Economics Jeanne Tschopp University of Bern Fall 2024 Motivation In earlier lectures, we developed an equilibrium labor market model in which there is no (involuntary) unem...
Topic 4.1: Unemployment Patterns Labour Economics Jeanne Tschopp University of Bern Fall 2024 Motivation In earlier lectures, we developed an equilibrium labor market model in which there is no (involuntary) unemployment Of course, this is not a realistic description of the world In this lecture we will overview a number of general facts about unemployment and discuss how job flows and worker flows can help us understand the observed unemployment patterns within and across countries Jeanne Tschopp Topic 4.1: Unemployment Patterns 1/ 43 Outline Look at the unemployment patterns in some industrialized countries Show how unemployment can be linked to job creation, job destruction, and worker flows into and out of unemployment Look at how the labor market is continuously reorganizing itself in most countries, with a large number of jobs being created while others are being destroyed, and a large number of workers losing their jobs while others are being hired Discuss the determination of an economy’s steady state unemployment rate Introduce the concept of the Beveridge curve which links the unemployment rate to the vacancy rate of the economy Reading: Cahuc et al textbook: Chapter 9.1 Jeanne Tschopp Topic 4.1: Unemployment Patterns 2/ 43 Defining Unemployment, Employment, and Participation Unemployment is characterized by 3 conditions (ILO guidelines) - persons who are without work - persons who are available for work - persons who are seeking work Unemployment rate: Ut 7 ut ⌘ , where L t = E t + Ut Lt Uz + 27 + 1 since thedenominator Employment rate: is not the same. Et et ⌘ where Pt = Lt + NLFt Pt Un+ ey1 Participation rate: ⌧t ⌘ Lt Since It> Lt Pt Notation - Ut : unemployment - Et : employment - Lt : workforce (labor force) - Pt : population - NLFt : not in the labor force Jeanne Tschopp Topic 4.1: Unemployment Patterns 3/ 43 Unemployment, employment, and participation Great diversity in the experiences of unemployment, employment and participation by countries It is generally observed that countries with high rates of employment and participation have lower rates of unemployment than others Evidence shows the existence of a decreasing relation between the unemployment rate and the - employment rate ) et is a relevant indicator of the abundance of jobs - participation rate Jeanne Tschopp Topic 4.1: Unemployment Patterns 4/ 43 Unemployment and Employment Rates - Decreasing relationship btw ut and ⌧t - Large heterogeneity across countries in ut and et - Spain and Grece (Austria, Norway, Switzerland): highest (lowest) ut - Switzerland exhibits the highest rate of employment and lowest rate of unemployment Jeanne Tschopp Topic 4.1: Unemployment Patterns 5/ 43 Unemployment and Participation Rates - Decreasing relationship btw ut and pt - ⌧t highly dispersed across countries (e.g. 51% in Turkey up to 82% in Switzerland) Jeanne Tschopp Topic 4.1: Unemployment Patterns 6/ 43 Changes in the Unemployment Rate across Countries - ut has evolved differently in Japan, continental Europe and the United States - Europe: low until the 1970s, then rose steadily until the late 1990s - Japan: stable ut until the 1990s, rose btw 1994-2001 - Oil shocks: ’74 and ’79; great recession: ’08 Jeanne Tschopp Topic 4.1: Unemployment Patterns 7/ 43 Simple accounting 5 buz [Sfopuahougo t excerate The Sources of Changes in the Unemployment Rate · · growth Ot The unemployment rate is defined by ut ⌘ Ut /(Et + Ut ). Thus, we have: ③ Employment i Et Pt ⌧ t = growth 1 ut D. E Taking this equation in logarithm at dates t and t 1, and using the approximation Lim ln z = z 1, we get: z!1 Pt ⌧t Et ut + = + Pt 1 ⌧t 1 Et 1 1 ut 1 We can deduce the variations in the unemployment rate (assuming that the ut is a small number): Pt ⌧t Et ut ' + Pt 1 ⌧t 1 Et 1 Jeanne Tschopp Topic 4.1: Unemployment Patterns 8/ 43 Et 1f T = 1 Ut- In It + Inte = InEc -In(1 -ut) AInEz + DInEt = DInEt-AIm(1-Ut & use the fact that Blue = P (from a first-order Taylor hear approx around (E-1 : -He In InCEDEE) Du T DAE + Art = DEAR O 1 - Uz- A Deempt a o =Aut = D z + EEl since population growth participlate growth Variations in the unemployment rate can come from changes in: 1 The number of employed people 2 The size of the working-age population 3 The labor force participation rate Note that it is entirely possible for the unemployment rate to fall without a rise in employment. - This may happen for instance if the labor force shrinks (either because the participation rate and/or the population goes down) Jeanne Tschopp Topic 4.1: Unemployment Patterns 10/ 43 Participations Rates over Time across Countries - Sharp increase in the labor force participation rate in the U.S. until ’95 - Europe (except Germany): stagnation of the participation rate Jeanne Tschopp Topic 4.1: Unemployment Patterns 11/ 43 Employment over Time across Countries ! - Sharp increase in the employment in the U.S. up to ’08 - Europe: much more modest ) poor performance in job creation (unlike the U.S.), leading to low employment growth - During last 30 years, European countries present a structural incapacity to create enough jobs. The weakness of job creation has been reflected in the rise of unemployment level Jeanne Tschopp Topic 4.1: Unemployment Patterns 12/ 43 Job and Worker Flows The labor market is characterized by intense reallocation of jobs and workers Job and worker flows: - play an important role in the determination of unemployment - represent a non-negligible fraction of the labor force For the period 1996-2003, around 15 million people changed labor market states or jobs in one month in the US, which corresponds to around 8% of the working age population Jeanne Tschopp Topic 4.1: Unemployment Patterns 13/ 43 8% of the working - age pop 2 6% experiencing. 2.6% hazard rate (h) 1.6%2.8 million of the persons population, (m) (p) 16-64 Kalbland - 2.6% of job-to-job switches - 1.3% of job-to-unemployment switches - 2.7% of job-to-out-of-the-labor-force switches Jeanne Tschopp Topic 4.1: Unemployment Patterns 14/ 43 Link Between Worker Flows and Employment Changes Net variations of employment are equal to the difference between workers’ entries into and exits out of employment It is also equal to the difference between job creation and job destruction: Net employment change = Hirings Separations | {z } Worker flows = Job Creations Job Destructions | {z } Job flows Jeanne Tschopp Topic 4.1: Unemployment Patterns 15/ 43 Hirings and Separations New hires may be coming from: - Unemployment or non-participation (“jobless-to-job”): inflows to employment - From another job (“job-to-job”): flows within the employment category Separations occur when individuals leave their job either because they are fired, quit, or retire, or because their fixed-term contract ends. They may leave to go to: - Unemployment or non-participation (“job-to-jobless”): outflows from employment - To a new job (“job-to-job”): flows within the employment category Jeanne Tschopp Topic 4.1: Unemployment Patterns 16/ 43 Employment inflows and outflows, annual rates, in % of employment stock, 2011 Inflows Outflows Country Entry rate (hirings) Exit rate (separations) Australia 23 24 Austria 16 16 Belgium 14 12 Canada 20 18 Czech Republic 12 11 Denmark 22 22 Finland 22 21 France 14 14 Germany 16 14 Greece 9 16 Hungary 14 13 Ireland 13 14 Italy 11 10 Korea 36 33 Mexico 26 25 Netherlands 17 4 Norway 16 15 Poland 14 13 Portugal 15 15 Slovak Republic 10 8 Spain 17 17 Sweden 21 18 Switzerland 18 16 Turkey 35 27 United Kingdom 15 14 United States 19 22 European Union (15 countries) 15 14 OECD (30 countries) 18 18 Source: OECD Labor Force Statistics Database. - Workers reallocations are important (generally larger than job reallocations, as we will see) - Worker mobility differs from country to country Jeanne Tschopp Topic 4.1: Unemployment Patterns 17/ 43 Job-to-job, Job-to-jobless and Jobless-to-job Flows There are important ! (but often not considered M basic search E / search-and matchinga - · job-to-job 1 job-bo-jobles - About half of entry-and-exit flows from employment involves job-to-job mobility without spells of unemployment Jeanne Tschopp Topic 4.1: Unemployment Patterns 18/ 43 Entry and Exit Rates into/from Unemployment · Bahre correlcon high inflows are asociated with high outflow ray · influ rates outflow rates · US us El - The rates of exit from unemployment are much higher than the rates of entry into unemployment, since average duration of unemployment is much shorter than the duration of spells in employment - Large heterogeneity across countires - Continental European labor markets appears to be sclerotic, to the extent that they display much lower rates of reallocation of labor Jeanne Tschopp Topic 4.1: Unemployment Patterns 19/ 43 The Role of Entry/Exit Rates in Unemployment Rate Variations Increases in unemployment may be driven by: - Higher flows into unemployment (more people becoming unemployed), and/or - Lower flows out of unemployment (fewer unemployed people becoming employed or non-participants) Researchers have studied the relative importance of these two channels - Elsby at al. (2013) find that variation in the outflow rate from unemployment explains 85% of overall variation in the unemployment rate for anglophone countries - Inflow rate explains only 15% - More even split for continental European and Nordic countries: 55% accounted for by outflows; 45% by inflows Jeanne Tschopp Topic 4.1: Unemployment Patterns 20/ 43 Hires over the Course of the Business Cycle in the U.S. - Hires are generally procyclical (like job creations): they decline during recessions Jeanne Tschopp Topic 4.1: Unemployment Patterns 21/ 43 Separations over the Course of the Business Cycle in the U.S. - Layoffs are countercyclical: they increase during recessions as firms tend to lay more employees off when the economy declines - (Voluntary) quits are procyclical: they decrease during recessions as wage earners have fewer job opportunities during downturns Jeanne Tschopp Topic 4.1: Unemployment Patterns 22/ 43 Job Flows We can also think of changes in employment from the perspective of jobs rather than workers Net employment change = Hirings Separations | {z } Worker flows = Job Creations Job Destructions | {z } Job flows Jeanne Tschopp Topic 4.1: Unemployment Patterns 23/ 43 Job creation and destruction Job creation represents the sum of job gains measured at the firm level (or establishment level according to the studies) over one year, due to the opening of new production units and the expansion of jobs within existing workplaces also creation of firms new Job destruction represents the sum of job losses resulting from the closing of production units and contractions in the number of jobs in units Note that worker flows do not necessarily entail job flows - A job may stay constant (no job flow) but different workers may go in and out of it (worker flows) Jeanne Tschopp Topic 4.1: Unemployment Patterns 24/ 43 Job reallocation is equal to the sum of job creation and job destruction Net employment growth is equal to the difference between these two quantities Excess job reallocation is the difference between job reallocation and the absolute value of net employment growth - reflects the amount of flows exceeding net employment growth - part of job reallocation exceeding the amount required to accommodate the net employment change Jeanne Tschopp Topic 4.1: Unemployment Patterns 25/ 43 Job Reallocation, Net Employment Growth and Excess Job Reallocation (2) (2) (3) (4) (5) Job Job Job Net Excess job Country (period) creation destruction reallocation Employment reallocation Argentina (95-02) 12.7 10.7 23.4 2.0 21.4 Brazil⇤ (96-01) 16.1 12.9 29.0 3.2 25.8 Chile⇤ (79-99) 11.6 11.3 22.8.3 22.5 Colombia⇤ (82-98) 10.5 10.0 20.5.5 20.0 Estonia (95-01) 13.3 12.0 25.3 1.3 24.0 Finland (88-98) 13.8 14.0 27.8 -.2 27.6 France (99-00) 12.0 8.3 20.3 3.7 16.6 Germany (77-99) 8.4 7.1 15.5 1.3 14.2 Hungary (92-01) 13.3 11.2 24.5 2.1 22.4 Italy (86-94) 12.3 10.2 22.5 2.1 20.4 Latvia (96-02) 15.7 10.8 26.5 4.9 21.6 Mexico (85-01) 16.9 12.0 28.9 4.9 24.0 Portugal (83-98) 12.5 10.7 23.3 1.8 21.4 Slovenia (92-01) 9.0 8.1 17.1.9 16.2 United Kingdom⇤ (80-98) 11.5 12.6 24.2 -1.1 23.1 United States (88-97) 12.5 10.0 22.5 2.5 20.0 ↓ Table: Job creation and destruction flows. = () (2) + = Annual average rate as a percentage of total employment, all sectors of the economy. 1 (2) : 15 - 1941/ Note: * Manufacturing only for Brazil, Chile, Colombia, and the U.K.. West Germany for Germany. Mostly private sector for Germany, Portugal, and the U.S. Source: data from Haltiwanger et al. (2010), see their table A.1 for exact sources; except for France where data is from Picart (2008, table 2) Jeanne Tschopp Topic 4.1: Unemployment Patterns 26/ 43 Job Reallocation, Net Employment Growth and Excess Job Reallocation (con’t) The Table shows wide variations in job creation and destruction flows Countries with low excess reallocation have lower shares of temporary employment, while low employment protection legislation (EPL) tends to be associated with higher reallocation rates For all countries, net employment growth is always much smaller than job creation or destruction Jeanne Tschopp Topic 4.1: Unemployment Patterns 27/ 43 Job Reallocation, Net Employment Growth and Excess Job Reallocation (con’t) These job creation and destruction figures do not include job reallocations that take place within individual firms (for example a firm that gets rid of an assembly-line worker and creates a managerial job) Studies that have attempted to assess job reallocations within workplaces suggest that this factor is not negligible - Hamermesh et al. (1996) find that reorganizations within firms explain 11% of overall job reallocations - Lagarde et al. (1995) estimate that job reallocations within firms in France represent almost half of all job reallocations Job movements most frequently take place within the same sector, rather than between different sectors (industries); high reallocation is not driven by changing industrial composition Jeanne Tschopp Topic 4.1: Unemployment Patterns 28/ 43 Job creation and destruction over time - Job destruction is countercyclical: it rises during recessions - Job creation is procyclical: it falls during recessions - Variations in job creation/destruction is strong during phases of recessions, and quite weak otherwise, implying that job reallocation is countercyclical Jeanne Tschopp Topic 4.1: Unemployment Patterns 29/ 43 9 Link Between Flow Rates and Unemployment Rates We get there using 2 eavations ~ g + f that descabe the D in stock of unemployed Assume that the size of the labor force is constant (i.e. no inflows or outflows to/from non-participation) Let qt denote the separation rate between period t and t + 1, and Et the total number of employed workers at time t ) Between periods t and t + 1, qt Et workers flow into unemployment Let ft denote the job finding rate, i.e. the probability that an unemployed worker finds a job between periods t and t + 1, and Ut the total number of unemployed people at time t ) Between periods t and t + 1, ft Ut workers will flow from unemployment into twin employment ona ou Jeanne Tschopp Topic 4.1: Unemployment Patterns 30/ 43 Link Between Flow Rates and the Unemployment Rate (con’t) ② Related to the flow rates Change in the unemployment stock: The change in the unemployment stock between periods t and t + 1, denoted as Ut+1 , is equal to the flows into unemployment minus the flows out of unemployment: Ut+1 = qt Et ft Ut (1) su a the mempl. # of a eining #of a pool finding job a and leaving the employment pool A pining the mempl pool. Jeanne Tschopp Topic 4.1: Unemployment Patterns 31/ 43 Link Between Flow Rates and the Unemployment Rate (con’t) ② Related to the def. In the mempl. rate The unemployment rate ut is defined as: ut ⌘ Ut /Lt , where Lt is the total labor force at time t. E) ve Ut = It The change in unemployment can therefore be written as: Ut+1 = ut+1 Lt+1 ut Lt Given the assumption that the size of the labor force is constant, we have that Lt+1 = Lt = L, so: Ut+1 = ut+1 L ut L Ut+1 = ut+1 L (2) Jeanne Tschopp Topic 4.1: Unemployment Patterns 32/ 43 Link Between Flow Rates and the Unemployment Rate (con’t) Combining (1) and (2) we have: ut+1 L = qt Et ft Ut all (3) 1 ut+1 = (qt Et f t Ut ) L (4) ut+1 = qt (1 ut ) ft u t (5) This equation shows how changes in the unemployment rate are a function of the flow rates and the current unemployment rate Jeanne Tschopp Topic 4.1: Unemployment Patterns 33/ 43 Link Between Flow Rates and Steady State Unemployment Rate The steady state is an equilibrium state of the economy where key variables remain constant The steady state unemployment rate is the long-run unemployment rate that the economy will converge to if the flow rates remain constant At the steady state, ut+1 = 0 ) ut+1 = ut = u⇤ , where u⇤ is the steady state unemployment rate que-fut o g(u)-f = = Based on Equation (5), if we set ut+1 = 0 and hold flow rates constant, we 19 can solve for u⇤ , the steady state unemployment rate: % su : q u⇤ = q q+f & bug - : - Decreasing in the job finding rate f - Increasing in the separation rate q Jeanne Tschopp Topic 4.1: Unemployment Patterns 34/ 43 The Beveridge Curve The evidence we have looked at suggests that the labor market is continuously reorganizing itself At every moment, a large number of jobs are being created and others are being destroyed; a large number of workers are losing their jobs and others are being hired Beveridge (1944) proposed to use the relationship between vacant jobs and the unemployment level to assess the extent of worker reallocation - Problems of reallocation ought to be greater, the higher the number of jobs vacant for a given number of unemployed Jeanne Tschopp Topic 4.1: Unemployment Patterns 35/ 43 The Beveridge Curve (con’t) -F unfilled positions Avacancies = # jobs ↓ + # unfilled positions Afilled positions - ↓ there are positions for more unfilled given i a 6 unempl - rate , compared to BC = 1 L The Beveridge Curve illustrates this link between the unemployment rate u and the vacancy rate v It illustrates the simultaneous presence of unemployed persons and vacant jobs Jeanne Tschopp Topic 4.1: Unemployment Patterns 36/ 43 US Beveridge Curve – Diamond and Sahin (2014) Jeanne Tschopp Topic 4.1: Unemployment Patterns 37/ 43 The Beveridge Curve in the United States The figure shows that the unemployment rate rose and the vacancy rate fell during the Great Recession in 2008-2009 During the recovery phase, more vacant jobs than before were required to reach the same unemployment rate. Thus, it seems that the BC has shifted outward Jeanne Tschopp Topic 4.1: Unemployment Patterns 38/ 43 Outward Shifts of the Beveridge Curve Outward shifts of the Beveridge Curve are interpreted as reflecting a reduced efficiency of the matching of worker to jobs: More difficult for firms and workers to meet due, for example, to information or geographic frictions Mismatch between characteristics of unemployed workers and characteristics that employers seek Jeanne Tschopp Topic 4.1: Unemployment Patterns 39/ 43 The Beveridge Curve in Germany · na Jeanne Tschopp Topic 4.1: Unemployment Patterns 40/ 43 The Beveridge Curve in Germany (con’t) Contrasting image to the U.S. as in Germany, the BC shifted to the left It appears that the fall in the unemployment rate is linked to an improvement in the matching proces between job seekers and vacant jobs (“German Miracle”) Possible reason: Hartz reforms (since 2003) - Eased regulations regarding layoffs - Stiffened regulations governing access to unemployment benefits - Benefits recipients had to accept jobs that do not correspond to qualifications or involve commuting - Resort on low-paid and short-term jobs (mini- and midi-jobs) Other reasons: - Short-time work: firm can get government subsidies to replace 2/3 of workers’ wages if it cuts the hours of its workers instead of laying them off - Working time accounts: have an employee work overtime without bonus but bank the overtime hours in the workers’ account that can be redeemed in the form of time off during a slack period Jeanne Tschopp Topic 4.1: Unemployment Patterns 41/ 43 Recap Unemployment rates are never zero, as predicted by neoclassical models; they tend to range between 5% and 10% in many developed countries In most countries, job creation and job destruction are large-scale phenomena - Total of these two flows amounts to between 15% and 30% of overall employment every year - Net employment growth is always much smaller than job creation or destruction Worker flows are systematically greater in size than job flows Jeanne Tschopp Topic 4.1: Unemployment Patterns 42/ 43 Recap We can understand how the unemployment rate is changing by considering changes in job creation and job destruction, as well as changes in the flows of workers into and out of unemployment An economy’s steady state unemployment rate depends only on its separation and job finding rates Vacant jobs and unemployed people co-exist at any point in time in all countries The Beveridge curve relates the unemployment rate to the vacancy rate Next step: Develop a model that features unemployment, based on workers’ job search behavior and firms’ job posting behavior Jeanne Tschopp Topic 4.1: Unemployment Patterns 43/ 43