Barriers to Entry, Deregulation and Workplace Training *

Transcription

1 Barrers to Entry, Deregulaton and Workplace Tranng * by Andrea Bassann (OECD and ERMES, Pars 2) Gorgo Brunello (Padova, CESfo and IZA) Prelmnary draft Abstract We develop a theoretcal and emprcal analyss of the mpact of barrers to entry on workplace tranng. Our theoretcal model ndcates that there are two contrastng effects of deregulaton on tranng. On the one hand, as stressed n the lterature, wth a gven number of frms profts per unt of output fall, reducng the sze of rents frms can enjoy f they tran ther employees. On the other hand, the number of frms ncreases followng deregulaton, thereby rasng the output gan from tranng and mprovng nvestment ncentves. Whle the balance of these effects s uncertan, numercal smulatons show that for reasonable values of the parameters a postve relatonshp between deregulaton and tranng prevals. We use repeated cross secton data from the European Labour Force Survey to nvestgate emprcally the relatonshp between product market regulaton and tranng ncdence n a sample of 5 European countres and 2 ndustral sectors, whch we follow for about 8 years. Our emprcal results are unambguous and show that a 0 percent ncrease n product market deregulaton ncreases tranng ncdence n the exposed ndustres by 2.8 to 5 percent, dependng on the selected emprcal specfcaton. Key words: tranng, product market competton, Europe JEL codes: J24, L * The vews expressed here cannot be attrbuted to the OECD or ts member countres. We are grateful to Alex Hjzen and Laura Nn Varta for ther help wth the data and to Kenn Arga, Georges Bresson, Mara De Paola, Fath Fakhfakh, Marco Leonard, Alfonso Rosola, Francsco Serranto, Sad Souam, Katsuya Tak, Govanna Vallant and to the audences n Gerona, Mlan Cattolca, Munch (CESfo), Osaka (OSIPP), ERMES and Pars 3 (CEPN) for comments and suggestons. The usual dsclamer apples. Address for correspondence : OECD, 2 rue André Pascal, Pars Cedex 6, France; Ph.: e-mal: andrea.bassann@oecd.org. Address for correspondence : Department of Economcs, Unversty of Padova, va del Santo 33, 3500 Padova, Italy; Ph.: ; e-mal: gorgo.brunello@unpd.t.

2 Introducton Product and labour market deregulaton have attracted consderable attenton by economsts and polcy makers, and there s a broad concern that these regulatons can hamper growth and ncrease unemployment (see Blanchard and Gavazz, 2003, Alesna et al, 2005). Emprcal research on the economc effects of product market regulaton and deregulaton has focused so far on employment (Kugler and Pca, 2004), productvty growth (Ncolett and Scarpetta, 2003, Aghon et al. 2006), nvestment n captal stock (Alesna et al, 2005), nnovaton (Aghon et al., 2005), and easness to start a busness (Djankov et al, 200). Ths paper develops a theoretcal and emprcal analyss of the relatonshp between product market regulaton / deregulaton and workplace tranng, whch we defne as the accumulaton of human captal takng place whle n employment and after school completon. We beleve that lookng at tranng s mportant because the producton of sklls s perceved as one of the man factors affectng productvty growth (see e.g. OECD, 2007). A casual look at avalable ndustry-level cross-country data reveals that the degree of strngency of product market regulaton and the tranng partcpaton rate are negatvely correlated. Usng data from the European Labour Force Survey and the OECD regulatory database whch cover the years , Fgure llustrates ths for 5 European countres and 3 non-manufacturng sectors (energy, transports and communcaton). It turns out that the regulaton ndex accounts for about 50 percent of the total varance of tranng. Yet, the prevalng vew n the labour economcs lterature s that product market competton depresses tranng because t compresses the sze of the rents that can be approprated by frms whch nvest n human captal and pay for most of job-related tranng (see e.g. Acemoglu and Pschke, 999, Gersbach and Schmutzler, 200, 2006) 2. Tranng ncdence s defned here as tranng partcpaton rate n the 4 weeks precedng the survey of full-tme employees aged between 25 and 54 years and wth at least 4 weeks of tenure. The OECD ndcator used here covers all aspects of antcompettve regulaton (excludng publc ownershp) and ranges between 0 no regulaton to 6 maxmum regulaton. The sector average of each varable has been subtracted to make comparable n a sngle charts fgures of dfferent ndustres. 2 The only excepton s Autor (200) who studes frm-sponsored tranng n the Temporary Help Sevce (THS) ndustry. In hs model tranng s provded at the tme of hrng and s essentally used by THS frms as a screenng and selecton devce. Greater product market competton ncreases the pressure to secure hgh-ablty workers and the need to use tranng as a way to attract and select them. Hs model s approprate for the THS ndustry, where yearly turnover rates are above 300% and therefore screenng s a major problem. It can also be generalzed to explan the relatonshp between product market competton and nducton tranng that s tranng provded at the tme of hrng n any ndustry. Yet, t cannot provde a general theory of the relatonshp between product market competton and tranng nsofar as t has no bearng on contnuous tranng that s tranng provded after hrng whch appears to be a large share of total frm-sponsored tranng (see Bassann et al et al., for a survey). 2

3 Fgure. The correlaton between tranng and regulaton tranng regulaton Note: The tranng partcpaton rate refers to tranng taken n the 4 weeks precedng the survey of full-tme employees aged between 25 and 54 years and wth at least 4 weeks of tenure. The regulaton ndcator covers all aspects of ant-compettve regulaton (excludng publc ownershp) and ranges between 0 no regulaton to 6 maxmum regulaton. The sector average of each varable has been subtracted to make comparable n a sngle charts fgures of dfferent ndustres. Data refer to averages for for 3 -dgt sectors (Energy, Transport and Communcatons) and 5 countres (Norway and pre-enlargement EU excludng Luxembourg). Sources: OECD regulatory database and Eurostat's European Labour Force Surveys. Ths vew also appears to be n contrast wth a standard fndng n the theory of ndustral organzaton: a frm s ncentves to reduce unt costs do not depend on the sze rents per se but on the senstvty of rents to unt cost reductons (see e.g. Boone, 2000, and Aghon et al., 997). Rent sze can matter only ndrectly, when t affects the response of rents to unt costs reductons. In partcular, n the case of prce competton (that s when frms set prces rather than quanttes), f frms have smlar producton costs, t s well known that greater competton ncreases ncentves to reduce unt costs even f rents defned n terms of profts per unt of output fall (see e.g. Aghon et al., 200). Ths occurs because reducng costs allows reducng prces and the senstvty of product demand to relatve prces s greater, the greater the degree of competton. Greater output gans from cost reducton appear to outwegh the loss due to compresson of unt rents. In ths paper we argue that a smlar mechansm apples to the nvestment n human captal by 3

4 frms, nsofar as frms pay for tranng only as a mean to reduce unt costs (see e.g. Stevens, 996, Acemoglu and Pschke, 999). We develop a model whch casts the tranng decson n an economy characterzed by mperfectly compettve product and labour markets where frms are ex-ante dentcal and compete n the product market by settng prces, as n Blanchard and Gavazz (2003). We show that n equlbrum deregulaton has two contrastng effects on tranng. On the one hand, a reducton n the barrers to entry for a gven number of frms compresses profts per unt of output, and thereby tends to depress tranng. Ths s the mechansm usually stressed n the lterature. On the other hand, and condtonal on profts per unt of output, addtonal frm entry due to deregulaton ncreases the output gans from tranng, thereby rasng the ncentve to nvest n tranng. Output gans ncrease because addtonal tranng reduces the relatve product prce and the response of output to prces s greater, the greater the degree of competton n the product market. We use a numercal smulaton to show that for reasonable values of the parameters the latter effect prevals on the former. Our focus on the tranng pad and organzed by frms s motvated by the fact that ths s the bulk of workplace tranng (see Bassann et al., 2007, for a survey on studes of tranng fnancng). We show, however, that the postve relatonshp between deregulaton and tranng holds also when t s the employee who nvests and bears the tranng costs: snce addtonal frm entry due to deregulaton ncreases the output gans from employng traned workers, the demand for sklled workers mproves, and so does ther supply. Emprcal work on product market competton and tranng s scarce. We are aware of only two studes on the ssue and both pont to a postve relatonshp between competton and tranng. Autor, 200, shows that the Herfndhal ndex and tranng are negatvely correlated n US temporary help frms, whch suggests that less competton reduces tranng. In a mult-country study of Europe, whch explots cross country and tme seres varatons n product market and labour market nsttutons, Bassann et al, 2007, fnd evdence of a negatve and statstcally sgnfcant correlaton between the ndex of product market regulaton developed by the OECD and tranng ntensty, thereby supportng and generalzng Autor's fndngs. One potental drawback of mult-country studes s that the combnaton of the country and tme dmensons does not allow to control for many confoundng factors that mght affect tranng and that vary across country and over tme. One way to overcome ths lmt s to add an addtonal dmenson to the data, as we do n the current paper. We buld on the approach followed by Bassann et al, 2007, by addng the sector as such dmenson. Wth data whch vary by country, year and sector, we can compare a treatment group, whch conssts of the sectors drectly affected by our measure of 4

5 deregulaton, wth a control group, composed of those sectors whch are not affected drectly by such measure, whle at the same tme controllng for country and country by tme specfc effects. The assgnment to sectors s random nsofar as we can safely rule out the possblty that frms antcpatng polcy reform can swtch sector of producton. Gven the costs of swtchng sector, we beleve that ths source of endogenety s of second order of relevance. We use the OECD database on product market regulaton whch has been desgned to pck up regulatory reforms n tradtonally heavly regulated sectors, such as transport, communcaton and publc utltes. We consder these sectors as our treatment group, and contrast ther behavour wth respect to tranng wth the control group, whch conssts of the manufacturng sector, lmtng our analyss to European countres. After the mplementaton of the European Sngle Market Programme (SMP hereafter) n the early 990s, we can assume that no sector-specfc regulatory reform affected the manufacturng sector, whch therefore qualfes as a genune control. 3 By comparng tranng ncdence n these sectors, we try to dsentangle the effects of deregulaton on tranng from other confoundng factors, whch affect both groups. To do so we match regulaton data wth tranng data from the European Labour Force Survey (ELFS) as well as several other sector-level databases, to obtan a rch sector-level database of tranng covarates. Our econometrc analyss confrms that reducng product market regulaton ncreases tranng ncdence. The estmated effects are sgnfcant. Dependng on the selected measure of regulaton and on the emprcal specfcaton, a 0 percent reducton of regulaton ncreases tranng ncdence n the exposed ndustres by 2.8 to 5 percent, whch suggests that the postve effect on tranng of the hgher number of frms prevals on the negatve effect assocated to the reducton of avalable rents. One mght queston whether the OECD ndcators can adequately capture the regulaton and deregulaton processes takng place n the treated ndustres. If yes, we would expect these ndcators to postvely affect ndustry specfc proftablty. We show that ths s the case when proftablty s measured by the observed Lerner ndex (defned as the average share of the value of output that exceeds varable costs, see e.g. Klette, 999). We also decompose the margnal effect of regulaton on tranng ncdence nto ts effect on the Lerner ndex and the margnal effect of the Lerner ndex on tranng, and show that the former s postve and the latter s negatve hgher profts are assocated to hgher rents but also to a lower number of frms, and the overall effect on tranng s negatve. The paper s organzed as follows. In the frst secton we develop the theoretcal model. The 3 The manufacturng sector, however, could be affected ndrectly, because of the close lnkage among sectors typcal of modern economes. 5

6 subsequent sectons ntroduce the emprcal strategy and the data and present our estmates of the relatonshp between tranng and product market regulaton. Conclusons follow.. The Model. Baselne model wth frm-specfc tranng Followng Blanchard and Gavazz, 2003, and Stevens, 996, consder a two - stage model economy where each frm produces a dfferentated product usng labour. The number of frms m s determned by an entry condton, whch s affected by product market regulaton. The logcal sequence of the model s as follows: frst of all, frms decde entry. In the frst preparatory stage after entry, each frm nvests n tranng and pays the tranng costs. In ths sub-secton we assume that the sklls mparted by tranng are frm specfc 4. Ths assumpton wll be relaxed n the next sub-secton. In both the decson to enter and n the choce of how many workers to tran, the employer can perfectly foresee the wages she wll pay n the fnal stage, the level of employment L, the prces she wll be able to sell her goods at and the number of frms operatng n the market. In the second and fnal stage, and condtonal on tranng, frms and workers bargan over wages and employment, prces are set and producton occurs. Frms n ths economy share the same producton and tranng technology, and the same elastcty of product demand. Rsk neutral workers have the same reservaton wage, and there are no exogenous separatons of workers from frms. We show that the equlbrum s symmetrc: all frms choose the same tranng ncdence, each product prce s equal to the average prce, and each traned worker receves the same wage. Wth symmetry, n equlbrum there s no ex-post labour moblty and ex-post poachng. Frms operate the technology e Y = AL [] where Y s output, e L s employment n effcency unts, A s the productvty of labour and the subscrpt s for the sngle frm. Labour conssts of traned ( L T ) and untraned ( LU ) workers. Traned 4 The latter assumpton s less restrctve than one mght thnk nsofar as each frm mght have an dosyncratc combnaton of general sklls (see e.g. Lazear, 2003). 6

7 workers are sklled and untraned workers are unsklled. Snce traned workers are more productve than untraned workers, employment n effcency unts s L e = L + γl [2] U T where γ > s the relatve effcency of sklled labour 5. Lettng T be tranng ncdence and L employment, L = T L and the producton functon can be wrtten as 6 T [ + ( ) T ] Y = AL γ [3] Snce employment s set by barganng n the second stage of the game, the decson on how many workers to tran n the ntal perod s equvalent to the choce of tranng ncdence T. Defne net and gross profts as net and gross of tranng costs. The characterzaton of the equlbrum proceeds by backward nducton and starts wth the decson concernng wages and prces. We assume that untraned workers do not bargan but receve the common reservaton wage V (ths assumpton s lfted n the appendx). 7 Traned workers, however, bargan over wages and prces wth the frm. Ex-post barganng s justfed by the fact that, when frms nvest n frm specfc tranng, workers cannot commt on ex-ante wages but re-contract after the tranng nvestment has taken place (Malcolmson, 997). Let the wage of untraned workers be W U = V and the wage of traned workers be W T = W, wth W V. We characterze the barganng as a cooperatve Nash game, whch we solve by maxmzng [( W V ) T L ] + ( β )ln{ PY L ( W T + V ( T ))} β ln [4] wth respect to wages and employment. Ths s equvalent to assumng that the partes are nvolved n effcent barganng. An alternatve characterzaton would be the rght to manage model, where the employer retans authorty over employment determnaton. We follow Blanchard and Gavazz n the 5 Total employment s defned as L = LU + LT. 6 See Dearden, Reed and Van Reenen, Ths assumpton s consstent wth estmates of Cahuc et al. (2006) for France, who fnd no barganng power for unsklled and medum-sklled workers. 7

8 preference for effcent barganng, whch they argue has the advantage of capturng the possblty that frms are not operatng on ther demand for labour. However, we show n the Appendx that our key qualtatve results are not affected f we were to choose the rght to manage model. In eq. [4] the element n brackets s the gan from a postve settlement accrung to traned workers, who have lnear utlty functons. Ths gan s weghted wth the relatve barganng power of sklled labour, β, and s defned as the dfference between the settled wage and the reservaton wage, whch accrue to sklled labour n the event of falure to reach an agreement. By defnton, unsklled workers do not gan from the bargan, and are bound to ther reservaton wages. In the Appendx we show that our key results stll hold when we allow unsklled labour to actvely partcpate to the bargan. The element n braces s the gross nomnal proft from a postve settlement accrung to the employer 8. We assume that profts n the event of falure to settle are equal to zero 9. Notce that employment settng s equvalent to prce settng, because labour demand s a derved demand and each frm faces the followng product demand functon θ Y P Y = [5] m P ' where θ = σg( m) >, σ s a sutable constant and g > 0, Y s aggregate output, and P and P are the prce of product and the average prce. As n Blanchard and Gavazz, 2003, the absolute value of the elastcty of demand wth respect to the relatve prce θ s ncreasng n the number of frms. Ths follows from the fact that an ncrease n the number of products assocated to entry rases the elastcty of substtuton between products. The outcome of the bargan s [ + ( γ ) T ] + V ( T β ) T W = βp A [6a] P A[ + ( γ ) T ] θ = [ W T + V ( T )] θ + β [6b] 8 Profts are gross because tranng costs n the second stage are bygones. 9 Falure to settle dsrupts the producton process, so that no producton occurs and the partes separate. When tranng s frm specfc ths separaton mples that the accumulated sklls are lost. 8

9 9 Smple manpulaton of these two condtons yelds 0 ( ) [ ] T PA V P P + = γ θ θ [7] We next turn to the ntal preparatory perod, when employers decde how many workers to tran. Let the real cost of tranng per employee be 2 ) ( 2 T T c μ = [8] a convex ncreasng functon of tranng ncdence. Ths assumpton captures decreasng returns n the producton of human captal and rules out corner solutons for optmal tranng ncdence. Usng [5] and [6a] n the defnton of net real profts Π, 2 we obtan ( ) ( ) ( ) ( ) ( ) ( ) [ ] θ θ θ γ θ θ γ μ γ θ β T A P V m Y T A T T PA V = Π [9] The term n braces on the rght-hand sde represents net real profts per unt of output π, whle the product of all the other terms on the rght-hand sde represents output Y. Snce the margnal varaton of real profts as a result of addtonal tranng s gven by ( ) ( ) ( ) ( ) ( ) ( ) ( ) T Y T T A T Y T Y T Y T ) ( ) ( = + = Π γ γ θ π γ π γ γ μ π π [0] 0 Employment s gven by [ ] [ ] θ γ γ + = + = P P T m A Y T A Y L ) ( ) ( The assumpton of a quadratc s made for smplcty, and can be justfed by assumng that, nter-ala, workers are heterogeneous n terms of tranng costs, wth unformly dstrbuted tranng costs (see secton.2). The analyss below remans vald, however, for any weakly convex functon. 2 That s, nomnal profts dvded by the average prce.

10 the frst-order condton for a maxmum wth respect to T ( Π / T = 0 ) yelds 3 ( θ ) A( γ ) π = μt [] Tranng s ncreasng n net real profts per unt of output π and, condtonal on π, s decreasng n the tranng cost parameter μ and ncreasng n the dfference between the productvty of sklled and unsklled workers A ( γ ) as well as n the elastcty of substtuton θ. The latter effect comes from the fact that the output gans from tranng are ncreasng n θ. Ths occurs because tranng ncreases productvty, relatve prces are decreasng n productvty (see eq. [7]) and the response of output to relatve prces s greater, the hgher s θ. Notce that the postve relatonshp between proftablty π and tranng only holds for a gven number of frms and for gven θ. However, snce θ cannot be gven n the long run, but s negatvely related to proftablty, the overall relatonshp between π and T can take ether sgn. Substtutng out π n [] wth ts expresson n terms of T, V and other parameters we obtan 2 2 V + θ μ ± μ + 4μ( γ ) ( β ) * T = P 2 > 0 [2] μ( γ )( + θ ) Snce both V and the parameters γ, θ, μ and β do not vary across frms, t must be that * * T = T, mplyng that all frms select the same tranng ncdence. By mplcaton, equaton [7] mples that at the optmal level of tranng frms have the same relatve prce. Snce n general equlbrum we cannot have that all frms have a relatve prce above or below (see Blanchard and Gavazz, 2003), t must P be that P s equal to, whch yelds V P θ = A γ θ [ + ( ) T ] [3] where we drop hereafter the subscrpt. Usng [3], [5] and [6a] n the defnton of profts, and 3 The second order condtons for a maxmum hold. 0

11 mposng the general equlbrum condton on prces, the equlbrum number of frms s determned by the condton that net profts per unt of output π must be equal to the cost of entry per unt of output ρ β μt π = = ρ θ 2A 2 [ + ( γ ) T ] [4] β where s gross real profts per unt of output. Blanchard and Gavazz, 2003, use a smlar θ convenent specfcaton for the cost of entry. In the Appendx we show that our results stll hold when we specfy more conventonally the cost of entry as a fxed cost, rather than as a cost proportonal to output. Condtonal on tranng, a fall n the costs of entry ncreases the number of frms and θ. As a consequence, profts per unt of output fall untl the arbtrage condton [4] s satsfed. A reducton n ρ can be brought about by lberalsaton reforms lowerng regulatory barrers n the product market. For nstance, one can assume that ρ = f (R), where R measures the strngency of regulatory barrers and f / R > 0. Usng [4] nto [] to elmnate profts per unt of output we obtan ( θ ) ( γ ρ μ T = A ) [5] Equatons [4] and [5] descrbe two schedules n the ( T,θ ) plane: the schedule TT assocated to [5] mples a postve relatonshp between tranng ncdence and the number of frms 4 (see Fgure 2). The reason s that, condtonal on net profts per unt of output, the output gans from tranng are larger the larger the elastcty of substtuton between products (see eq. []). On the other hand, the schedule MM assocated to [4] suggests a negatve relatonshp between T and θ. Condtonal on net profts per unt of output, the greater the tranng, the greater the gross profts that are requred to avod losses. In turn, ths mples that, f net profts are fxed by entry barrers, the greater the tranng the smaller the number of frms that can survve n equlbrum wthout makng losses. We establsh the followng 4 θ Recall that ' = ( m) > 0 m σg.

12 Lemma. The long run equlbrum exsts and s unque f the followng two condtons hold: ρ β A( γ ) ρ + μ 2Aγ ( β ) A( γ ) ρ μ + 2Aγρ Proof: Equatons [4] and [5] descrbe two schedules n the ( T,θ ) plane: the schedule TT assocated to [5] has a postve slope and the schedule MM assocated to [4] has a negatve slope. Tranng ncdence tends to zero as θ n [5] and to a non-negatve number n [4] f ρ β. Snce ncdence ranges between 0 and, an equlbrum exsts f the two schedules ntersect for 0 T. T TT MM θ Fgure 2: The relatonshp between θ and T A( γ ) ρ + μ 2Aγ ( β ) The condton guarantees that ths s the case by mposng that at T = the A( γ ) ρ μ + 2Aγρ value of θ on the TT lne s at least as large as the value on the MM lne. When the two condtons are satsfed the two schedules ntersect once and only once n the relevant doman of the two varables. QED 2

13 We can now establsh the followng proposton concernng equlbrum comparatve statcs: Proposton. Subject to the condtons of the above lemma, a reducton n entry barrers whch corresponds to a reducton n the parameter ρ ncreases tranng ncdence f β θ ( ρ) β < ρ θ ( ρ) θ ( ρ) θ ( ρ) and reduces tranng ncdence f β θ ( ρ) 0 < ρ < θ ( ρ) θ ( ρ) [6] Proof: total dfferentaton of [4] and [5] yelds Σ dm + Σ 2dT = Σ 5dρ Σ 3dm + Σ 4dT = Σ6dρ where ' β ' Σ = σ g ( m)( γ ) ρa Σ 2 = μ Σ 3 = g ( m) 2 σg( m) Σ = ( σ g ( m) )( γ ) A Σ 5 6 = Σ 4 2μT + μt = 2A 2 ( γ ) [ + ( γ ) T ] 2 Snce the determnant of the Jacoban s negatve when θ < + and ρ > 0, dt dρ < 0 f Σ Σ6 Σ3Σ5 > 0, or β θ ( ρ) < ρ θ ( ρ) θ ( ρ) β Non-negatve profts also mply see [5] that ρ. Therefore dρ < 0 θ ( ρ) β θ ( ρ) β < ρ θ ( ρ) θ ( ρ) θ ( ρ) Clearly, dt > 0 dρ f β θ ( ρ) > θ ( ρ) θ ( ρ) ρ. QED We also establsh the followng corollary dt f 3

14 Corollary. A reducton n the barrers to entry ncreases the number of frms m. Proof. Follows from the proof above and from the fact that Σ Σ Σ Σ > In comparatve statcs terms, the effect on tranng ncdence of product market deregulaton brngng about a reducton n entry costs s graphcally llustrated n Fgure 3, where contnuous lnes refer to the ntal equlbrum and dashed lnes to the new equlbrum. Whle the effect of such a deregulaton on the number of frms s not ambguous, the effect on tranng depends on the relatve shft of the two curves. Takng as gven the tranng ncdence of each frm, a reducton n the barrers to entry ncreases the number of frms, thereby shftng the MM schedule outwards see equaton [4]. Ths, n turn, ncreases tranng ncdence along the TT schedule snce, condtonal on net profts, output gans from tranng are greater see equaton [5]. However, for a gven number of frms, lower entry barrers also rotate the TT schedule downwards, 5 because they reduce net profts per unt of output see equatons [0] and [5]. Ths reduces tranng ncdence. The overall effect on T s ambguous, but s postve f condton [6] holds. Condton [6] s less lkely to hold when entry costs are already low or the number of frms s hgh ( θ ). However, for reasonable values of the parameters t holds even for relatvely low values of ρ. To llustrate ths numercally, we assume a lnear relatonshp between the cost of entry and regulatory barrers, as measured by regulatory ndcators (see next secton). We then set the confguraton of parameters n such a way that, when the rato of the measure of barrers to entry to ts maxmum possble value ( β ) s equal to the observed medan rato n the sample for whch we have tranng and regulaton data (3 ndustres, , 5 countres, see Fgure above), we obtan approxmately an optmum tranng ncdence equal to the observed medan tranng ncdence. Ths rato s equal to 0.62 accordng to our ndcators. Assumng wth no loss of generalty that A =, we consder all possble combnatons of β, γ and μ such that for ρ = 0.62*( β ) we obtan that optmal tranng ncdence s equal to the medan level that we observe n our data (8.0%). 6 Wth ths choce of parameters, a deregulaton whch cuts ρ ncreases tranng for most of the range of varaton of ρ. For nstance, for any value β and realstc values of γ, say γ. 7 whch mples that the productvty of traned workers s no greater than 70% the productvty of untraned workers 5 The barycentre of the rotaton s the pont (,0). 6 Where tranng ncdence s defned n terms of the partcpaton rate as n Fgure. A smlar pcture (see appendx) can be obtaned f we consder the stock of tranng computed by perpetual nventory method (see next secton for the method of calculaton), whch corresponds better to the concept of tranng ncdence developed n the model. 4

15 deregulaton rases tranng f ρ /( β ) s at least as large as 0.46 (see Fgure 4), 7 a low threshold whch has rarely been attaned n hstorcal perspectve, and whch s not attaned by about 7% of the observatons for whch we have data. T TT TT MM MM θ Fgure 3: The effect of reducng barrers to entry Even more, dt 0 n a relatvely large regon, and t starts gettng sgnfcantly greater than dρ zero only for values close to ρ /( β ) = 0. 06, that corresponds to less than 3% of the observatons for whch we have data. Conversely, for values above 0.4, whch corresponds approxmately to the frst quartle n our data, the relatonshp between barrers to entry and tranng stocks s negatve and very close to lnear. 7 Fgure 4 llustrates ths for β = 0.5. For any other value of β n the nterval [0,), the admssble range of varaton of regulaton dffers. Yet, by rescalng the ρ-axs, one obtans very smlar equlbrum manfolds as n Fgure 4. 5

16 Fgure 4: The relatonshp between ρ and T wth β = 0.5,.05 γ.7 and μ such that medan tranng rates correspond to medan regulaton levels.2 General tranng: tranng versus hrng sklled labour In the economy descrbed so far, each frm nvests n frm specfc tranng and s wllng to pay the necessary tranng costs. Suppose now that sklls are at least partly general and that at the end of the ntal perod before the bargan occurs frms have the opton of hrng already sklled labour from the labour market or from other frms (poachng). If hrng were costless and hred workers were as productve as workers traned n-house, no frm would be wllng to tran and bear the tranng costs. However, hrng sklled labour n an mperfectly compettve labour market s costly, ether because sklls are not fully transferable or because of the presence of frctons and nformaton asymmetres. For 6

17 example, one can magne that hrng sklled labour requres a costly both n terms of tme and resources screenng actvty n order to dstngush between sklled and unsklled applcants and to measure the sklls of potental hres. Assumng that all frms have the same hrng costs, frms decde whether to tran or not by comparng margnal tranng and hrng costs. The results n Secton. hold when the margnal cost of tranng s lower than the margnal cost of hrng for any value of tranng ncdence. In ths case, frms wllngness to pay for tranng s justfed because labour market frctons ncludng search and nformatonal asymmetres substantally reduce the transferablty of general sklls from a frm to another, as dscussed by Acemoglu and Pschke (999). Consder now the opposte case, where hrng s less costly than tranng. Wth no heterogenety n the technology and n the costs of hrng and tranng, all frms wll prefer to hre sklled labour than tran unsklled labour. But wth no tranng by frms there can be no hrng of sklled labour, unless workers themselves are wllng to bear the tranng costs. We study ths case by assumng that rsk neutral workers who nvest n tranng acqure smlar sklls but pay an dosyncratc tranng cost ξ, whch s unformly dstrbuted n the populaton wth constant densty ψ. 8 Then the proporton of workers nvestng n tranng conssts of those ndvduals wth tranng costs that are lower than or equal to the expected real gan from tranng tranng supply curve s W V P, and the W V ψp = T [7] suggestng that tranng ncdence ncreases wth the sklled wage gap. When workers dffer only n ther tranng costs and tranng s general n nature, wages W are determned by the supply and demand for sklled labour. In the second stage each frm takes wages as gven and selects employment and the percentage of sklled employees to maxmze real profts 8 The results obtaned n the prevous subsecton for frm-specfc tranng and homogeneous tranng costs can be easly obtaned wth heterogeneous unformly dstrbuted costs f: ) frms do not know ndvdual tranng costs at the tme of hrng but they know the dstrbuton of these costs, so that by the law of large numbers each frm faces the same dstrbuton of tranng costs ndependently of ts sze; ) upon hrng tranng costs are revealed and frms frst tran workers wth lower costs. Under these assumptons the analyss of secton. can be replcated by smply replacng μ wth ψ. 7

18 Π = θ θ { AL [ + ( ) T ]} L T L ( T ) Y W V γ [8] m P P The frst order condton wth respect to tranng s P θ P W V AL ( γ ) = L ( ) P [9] Imposng the equlbrum condton P = P and usng [7] n [9] we obtan that tranng ncdence s gven by γ T = A [20] θ ψ The above equaton ndcates that, when workers nvest n tranng and sklls are general, a deregulaton of the product market whch rases the number of frms and the elastcty of product demand θ generates hgher tranng ncdence because t ncreases the margnal benefts of tranng accrung to frms and the demand for sklled labour. Hgher demand generates n turn hgher wages for sklled labour, whch nduce a hgher supply of sklls. We conclude that emprcal evdence showng that regulaton reduces tranng ncdence s consstent both wth an equlbrum dscussed n subsecton., where frms tran and bear the tranng costs, and wth an alternatve equlbrum where workers bear the tranng costs and frms hre sklled labour from the market. 2. The Emprcal Model In the emprcal applcaton, we are nterested n nvestgatng whether changes n regulaton affect tranng ncdence, measured as the proporton of workers recevng tranng. Equatons [4] and [5] generate n mplct form a map from entry costs ρ to tranng ncdence T, whch s plotted n Fgure 4 for a selected confguraton of parameters. To the extent that the relatonshp between entry costs ρ and the degree of strngency of regulatory barrers s monotonc, ths map s also from regulatory barrers R to tranng ncdence T. As descrbed n detal n the data secton below, we do not 8

19 dspose of a drect measure of entry costs. Yet, gven that our ultmate nterest s establshng an emprcal relatonshp between regulaton and tranng, we can estmate drectly the mplct equlbrum mappng between R and T brought about by the model. Alternatvely, snce n the long-run equlbrum entry costs are equal to (net) real profts per unt of output, we can use a measure of proftablty, such as the Lerner ndex, to proxy entry costs. In ths case, the relatonshp between deregulaton and tranng conssts of two components: the effect of proftablty on tranng and the effect of product market regulaton on proftablty (entry costs). However, the problem wth any measure of proftablty s that t s endogenous, as productvty shocks whch affect tranng also nfluence profts. Fortunately, and lettng asde poltcal economy consderatons, regulaton can be assumed to be exogenous and the degree of strngency of regulatory barrers can be used as an nstrument for the measure of proftablty. 9 Our regulatory ndcators and proftablty measures vary by sector, country and over tme. Therefore, we collapse our data on tranng and addtonal controls at the same level of aggregaton. As dscussed n the prevous secton, for most of our sample a monotone (negatve) relatonshp between entry costs (regulatory barrers) and tranng prevals. We can therefore estmate the followng emprcal counterpart of the theoretcal model: T ct = λ 0 + λ X ct + λ2yct + ε ct [2] where the vector X ncludes a vector of controls, such as average age, educaton and frm sze, Y s the selected measure of regulaton or proftablty, the subscrpt s for the ndustry, c s for the country and t s for tme. The lnear specfcaton n [2] can be problematc when the dependent varable s fractonal (see Wooldrdge, 999). Followng Papke and Wooldrdge (996), we assume that the condtonal mean s a logt functon G of the ndependent varables and consder the followng transformed generalzed lnear model (GLM) as an alternatve to (2): T ct ( λ + λ X ct + λ Yct ) + ε ct = G 0 2 [22] 9 Regulaton mght be endogenous f deregulaton tends to occur where productvty growth or nnovaton are greater and tranng s correlated wth these varables (see e.g. Duso and Röller, 2003). Although we beleve that ths s unlkely to 9

20 that we estmate usng a quas-maxmum lkelhood estmator (QMLE), where the quas-lkelhood functon s the bnary choce log lkelhood. 20 We postulate that the error term s as follows ε = ξ + ξ + ξ + ω [23] ct c t ct ct where ξ c s a country by ndustry effect, ξ t s an ndustry by tme effect, ξ ct s a country by tme effect, and ω ct s a standard dsturbance. We control for these unobserved effects by ncludng n the specfcaton country by sector, country by year, and sector by year dummes. The country by sector dummes capture cross country dfferences n the structure of each ndustry, ncludng dfferences n the parameters σ, γ, β and A; the sector by year effects capture the tme varyng dfferences n trend growth between affected ndustres (treated group) and ndustres not affected drectly by deregulaton (control group); the country by year dummes absorb country-specfc macroeconomc effects, countrywde changes n polcy (notably tranng polcy and naton-wde regulaton, on whch we have no data - see below) as well as changes n the routng of the questonnare and/or the exact formulaton of the tranng queston. Our key nterest les n estmatng the coeffcent λ 2, whch measures the relatonshp between product market regulaton and tranng. A negatve estmate of λ 2 would confrm that a negatve relatonshp between regulaton and tranng prevals. Strctly speakng, however, our theoretcal model suggests that ths coeffcent s postve for low levels of regulaton and negatve for hgher values. We can therefore try to capture ths bell-shaped pattern by fttng a quadratc n our regulatory ndcators. Yet, snce we do not have data on naton-wde aspects of regulaton affectng all sectors, we wll have to do a number of arbtrary assumptons to estmate these quadratc specfcatons, and results obtaned from them must be vewed as a tentatve extenson. One can argue that the tranng concept developed n the model corresponds to steady state tranng stocks. Yet, we only have data on tranng flows (see next secton). However, gven that serously affect estmates, we partally control for ths channel by addng to some of our specfcatons, as addtonal covarates, R&D ntensty, productvty levels and growth and the nvestment rate. 20 Papke and Wooldrdge (996) show that QMLE estmators of ths knd yeld consstent estmates of equaton [22] ndependently of any assumpton on the error term, for whch a robust varance estmator can be easly devsed. In addton, n contrast to the more classcal WLS estmaton of a lnear model wth log-odds transformaton of the dependent varable, the GLM specfcaton does not requre adjustment for boundary values (such as zeros) and can be estmated when fractonal data are obtaned by sample averages n samples of unknown sze that cannot therefore be used to construct weghts, as s our case. 20

21 tranng stocks and tranng flows are lkely to be postvely correlated, we can re-specfy our emprcal model n terms of the tranng flow τ, by smply substtutng t for T n equaton [2]. Alternatvely, we can construct tranng stocks from tranng flows by followng a methodology smlar to Cont (2005) and Dearden et al. (2006). In partcular, we assume a common deprecaton rate ( δ = 0. 5 ) and a steady state rate of tranng growth g equal to the sample average growth rate of tranng flows 2, and reconstruct ntal condtons under the assumpton that steady state growth occurs at the begnnng of the sample, whch mples τ /( δ + g ) = T0, where τ s the tranng flow n the frst perod and T 0 the tranng stock n the ntal perod. Fnally, tranng stocks after mssng years are constructed by assumng steady state growth n those years. Insofar as these data are reconstructed, we prefer to specfy our model n terms of flows and use tranng stock data only n a senstvty analyss. 22 Specfcatons [2] and [22] assume that, condtonal on the vector X, product market regulaton varables are the only nsttutonal varables that can vary by country, year and sector, and that nsttutonal confoundng factors are fully accounted for by the combnaton of country by year, country by sector and sector by year dummes. Whle ths s plausble, we cannot rule out the possblty that varables measurng labour market nsttutons at the same level of detal as regulaton varables could affect tranng ncdence. If ths s so, falure to account for these effects could erroneously attrbute them to changes n product market regulaton. Therefore, we also experment wth specfcatons that augment the vector X wth avalable measures of labour market nsttutons. 3. Data We use three man sources of data: a) an OECD database on tranng and other labour market varables; b) OECD regulatory ndcators for seven non-manufacturng ndustres (electrcty, gas, ar transport, road transport, ralways, post and telecommuncatons); c) addtonal sector-level nformaton on output, physcal captal etc ) avalable n the OECD STAN Famly databases and the companon 60-ndustry database of the Gronngen Growth and Development Centre. The OECD database on tranng s drawn from the EU Labour Force Surveys. It contans nformaton on tranng and a number labour market varables (namely age, gender, educaton, parttme/full-tme status, occupaton, ndustry, frm-sze, tenure, whether the contract s temporary or 2 In the steady state flows and stocks grow at the same rate. In country/sector unts where a decrease of tranng flows s observed, the steady state growth rate s set to zero. 2

22 permanent, whether the actvty s n the country of resdence, partcpaton n tranng, type of tranng and tranng duraton) for employed workers of 23 European countres from 995 to 2002 (wth many mssng values, correspondng to countres and years where questons on tranng were not admnstered or data on tranng are unrelable). Data have been collected n the second quarter of each year (March n most countres). Quanttatve varables (such as tenure or frm sze) are dvded nto categores (see the appendx for more detals). As regards to age, we dspose of categores coverng fve-year ntervals. We then reconstruct an ordnal varable by applyng to all observatons n each category the mean age of the category. Data are sem-aggregated nsofar as organzed by cells. Each cell corresponds to a combnaton of categores. Avalable cells cover all non-empty combnatons wth one category for each varable. Populaton weghts are reported for each cell. Tranng data refer to partcpaton n any educaton or tranng course n the 4 weeks precedng the ntervew ( week for France). Data on the type of tranng and ts length are often mssng. For ths reason, we do not use ths nformaton. In order to avod that ntal and close-to-retrement educaton confound the nformaton on workplace tranng, we lmt our analyss to full-tme employees wth at least month of tenure, aged between 25 and 54 years, and workng n ther country of resdence. Descrptve statstcs on tranng are avalable n the Appendx. We collapse our data on tranng and selected other labour market varables (educaton, age, gender and frm sze) at the level of sectors. Industres are avalable at the 2 dgt level of the NACE rev. classfcaton (that, at the 2 dgt level, corresponds to the ISIC rev. 3, wth extremely few exceptons) for most countres and years n manufacturng. In servces, they are avalable at a slghtly more dsaggregated level than dgt of NACE rev.. However, snce only NACE 970 s avalable for 992 and for certan country-year pars, we are oblged to aggregate a few ndustres n order to construct our sector-level database. The fnal lst of ndustres s avalable n the Appendx. We have access to detaled OECD ndcators on ant-compettve product market regulaton for seven 2 or 3 dgt non-manufacturng ndustres. Data are avalable on an annual bass and span from 975 to 2003 for 2 OECD countres. Followng Alesna et al. (2005) we use these data to construct tme-seres ndcators of regulatory barrers for three more aggregate ndustres (utltes, transport, and communcaton servces), for whch we have tranng data (see above). Detaled regulatory ndcators concern sector-specfc entry barrers, publc ownershp, the market share of the domnant player(s) when relevant (n the telephone, gas and ralroad sectors), vertcal ntegraton n network ndustres and 22 In the senstvty analyss we wll consder also three other measures of the tranng stock obtaned by varyng constructon hypotheses. 22

23 prce controls when relevant (n the road freght ndustry). Outsde the scope of these ndcators are naton-wde aspects of regulatons applyng to all ndustres, such as admnstratve barrers to entrepreneurshp (admnstratve barrers on start-ups, general features of the lcensng and permt system, etc...), snce these data are not avalable n tme-seres n the OECD database. Avalable ndcators vary between 0 and 6 from the least to the most regulated. Entry barrers cover legal lmtatons on the number of companes and rules on vertcal ntegraton of network ndustres. A value of 0 corresponds to free entry. By contrast, a value of 6 apples when entry s severely restrcted. Publc ownershp measures the share of equty owned by central or muncpal governments, and takes a value of 0 n the case of no equty and a value of 6 n the case of full ownershp. All other ndcators are smlarly defned. More detals on these ndcators are avalable from Ncolett et al. (200), Alesna et al. (2005) and Conway and Ncolett (2006). We construct our aggregate ndcators followng the same methodology of Alesna et al. (2005). Ths nvolves two steps. Frst, separate ndcators of barrers to entry, publc ownershp, market structure, vertcal ntegraton and prce controls for each of the seven ndustres are averaged to obtan two coarser (and partally alternatve) ndcators: BEVI, whch summarzes barrers to entry (comprsng legal restrctons and vertcal ntegraton) and REGNO, whch ncludes all dmensons except publc ownershp 23. As our model apples explctly to barrers to entry we expect BEVI to be related to tranng. The same mght apply to REGNO nsofar as the presence of prce controls and the absence of barrers to concentraton can, to a certan extent, be seen as addtonal barrers to start-ups. Second, the same ndcators for the three more aggregated ndustres are obtaned by smple averagng the values of the correspondng sub-ndustres. Once the two ndcators of regulaton are matched to our tranng data we obtan 32 country by sector by tme non-mssng observatons concernng three (typcally regulated) servce ndustres for 5 European countres and a maxmum of 8 years. 24 Yet, these ndustres account for a small share of total employment (about 7.5% n 2002). Moreover, n the event that reforms n these three sectors have occurred almost smultaneously, the effect of the varaton n regulaton on tranng ncdence rsks to be swept away once country per year dummes are ncluded n the emprcal analyss. 23 A thrd ndcator, REGPO, whch summarzes the degree of publc ownershp, s also consdered. Specfcatons augmented by ths ndcator are estmated as a robustness check. Ths ndcator however turns out nsgnfcant wthout affectng sgnfcantly the coeffcents of other co-varates. The lack of sgnfcance of ths ndcator s consstent wth our model where publc ownershp plays no role. In addton ths s also consstent wth the evdence that suggests that publc sector employees receve no less tranng than ther prvate sector counterparts (see e.g. Bassann et al;, 2007). 24 Snce the exact month of each regulatory reform s not known and mght well be subsequent to the second quarter of the correspondng year, each regulatory ndcator s lagged one year. 23

24 To crcumvent ths problem, we add manufacturng ndustres to our dataset. For these ndustres, regulaton concerns essentally admnstratve burdens, lmtatons to foregn drect nvestment as well as barrers to trade, at least n European countres. Now, only the last two barrers can be consdered to be sector-specfc. But for them, due to the comng nto acton of the Sngle Market Programme (SMP) n 992, t can be assumed that ther tme profle s flat snce at least 994 for the 2 countres that were EU members n 992 (see e.g. Bottasso and Sembenell, 200). 25 The same argument can be appled to Austra, Fnland, Norway and Sweden from 995 (see e.g. Baldwn et al., 996, and Gullstrand and Johansson, 2005). 26 Ths s equvalent to assume that snce 994 for the majorty of countres and snce 995 for a few countres - regulaton n manufacturng has been equal to an arbtrary constant, whch we control for wth country by ndustry dummes. In sum, we construct an extended dataset startng from the three non-manufacturng ndustres by addng manufacturng and by droppng pre-smp years. In practce, ths s equvalent to usng a dfference n dfference estmator, where manufacturng sectors are used as the control group and servces are the treated group. We end up wth a sample composed of 5 countres and 2 ndustres for a maxmum of 8 years for each country-ndustry par, and for a total of 236 observatons once observatons wth mssng nformaton on tranng s excluded. As dscussed above, deally we would lke to estmate a bell-shaped relatonshp between regulaton and tranng. However, strctly speakng ths s mpossble n our data, snce we do not have a tme-seres for those naton-wde aspects of regulaton that apply to all sectors (such as admnstratve barrers to start-ups). In specfcatons such as [2] ths component of regulaton s controlled for by country by tme dummes. However, ths s no longer the case n specfcatons wth a quadratc. On a tentatve bass we can add the naton-wde ndcator of admnstratve barrers to start-ups n 998 to our sector-specfc ndcators of regulaton, assumng the latter to be 0 n manufacturng, so as to obtan a reconstructed ndcator of regulatory barrers that s more comparable across countres. Yet ths ndcator wll mss the effect of admnstratve reforms that took place durng the years As hnted at above, we consder also the observed Lerner ndex as a measure of unt proftablty n real terms. In our data the Lerner ndex L s defned as the dfference between the value 25 Bottasso and Sembenell (200) report that, on average, 75% of the measures mpled by the SMP agreement were already transposed nto natonal legal systems at the tme when the SMP came nto acton, and that vrtually all measures were transposed shortly after. 26 Accordng to the frst edtons of the Sngle Market Scoreboard (EC, 997, 998), by 997 Fnland, Norway and Sweden were among the best performng countres as far as transposton of EC drectves s concerned. Only Austra appeared to lag behnd, but ts gap wth other EU countres was closed n 998. For ths reason we check that our results are robust to excluson of Austra pror to