Skip to main content


<?xml version="1.0"?>

<!DOCTYPE TEI.2 SYSTEM "base.dtd">




<title>Product market influences on wages</title></titleStmt>

<publicationStmt><distributor>BASE and Oxford Text Archive</distributor>


<availability><p>The British Academic Spoken English (BASE) corpus was developed at the

Universities of Warwick and Reading, under the directorship of Hilary Nesi

(Centre for English Language Teacher Education, Warwick) and Paul Thompson

(Department of Applied Linguistics, Reading), with funding from BALEAP,

EURALEX, the British Academy and the Arts and Humanities Research Board. The

original recordings are held at the Universities of Warwick and Reading, and

at the Oxford Text Archive and may be consulted by bona fide researchers

upon written application to any of the holding bodies.

The BASE corpus is freely available to researchers who agree to the

following conditions:</p>

<p>1. The recordings and transcriptions should not be modified in any


<p>2. The recordings and transcriptions should be used for research purposes

only; they should not be reproduced in teaching materials</p>

<p>3. The recordings and transcriptions should not be reproduced in full for

a wider audience/readership, although researchers are free to quote short

passages of text (up to 200 running words from any given speech event)</p>

<p>4. The corpus developers should be informed of all presentations or

publications arising from analysis of the corpus</p><p>

Researchers should acknowledge their use of the corpus using the following

form of words:

The recordings and transcriptions used in this study come from the British

Academic Spoken English (BASE) corpus, which was developed at the

Universities of Warwick and Reading under the directorship of Hilary Nesi

(Warwick) and Paul Thompson (Reading). Corpus development was assisted by

funding from the Universities of Warwick and Reading, BALEAP, EURALEX, the

British Academy and the Arts and Humanities Research Board. </p></availability>




<recording dur="00:43:05" n="7318">


<respStmt><name>BASE team</name>



<langUsage><language id="en">English</language>



<person id="nm0785" role="main speaker" n="n" sex="m"><p>nm0785, main speaker, non-student, male</p></person>

<person id="sm0786" role="participant" n="s" sex="m"><p>sm0786, participant, student, male</p></person>

<person id="sm0787" role="participant" n="s" sex="m"><p>sm0787, participant, student, male</p></person>

<person id="sf0788" role="participant" n="s" sex="f"><p>sf0788, participant, student, female</p></person>

<person id="sf0789" role="participant" n="s" sex="f"><p>sf0789, participant, student, female</p></person>

<person id="sm0790" role="participant" n="s" sex="m"><p>sm0790, participant, student, male</p></person>

<person id="om0791" role="observer" n="o" sex="m"><p>om0791, observer, observer, male</p></person>

<person id="sm0792" role="participant" n="s" sex="m"><p>sm0792, participant, student, male</p></person>

<person id="sm0793" role="participant" n="s" sex="m"><p>sm0793, participant, student, male</p></person>

<person id="sm0794" role="participant" n="s" sex="m"><p>sm0794, participant, student, male</p></person>

<person id="su0795" role="participant" n="s" sex="u"><p>su0795, participant, student, unknown sex</p></person>

<personGrp id="ss" role="audience" size="s"><p>ss, audience, small group </p></personGrp>

<personGrp id="sl" role="all" size="s"><p>sl, all, small group</p></personGrp>

<personGrp role="speakers" size="13"><p>number of speakers: 13</p></personGrp>





<item n="speechevent">Lecture</item>

<item n="acaddept">Economics</item>

<item n="acaddiv">ps</item>

<item n="partlevel">PG</item>

<item n="module">Foundations of Economic Analysis</item>




<u who="nm0785"> on this i've printed it off in a machine which was a clever machine 'cause it 'cause it's given you the <pause dur="0.8"/> the home page address on this <pause dur="0.3"/> thing as well <pause dur="0.6"/> okay so <pause dur="0.3"/> what i advise you to do <pause dur="0.3"/> is <pause dur="0.4"/> to <pause dur="0.5"/> go <pause dur="0.2"/> to the home page for this course to find it either straight from this address or just the department's home page <pause dur="0.3"/> my home page there's a link in there to this M-S-C <pause dur="0.9"/> labour market economics home page <pause dur="0.3"/> okay <pause dur="0.5"/> so <pause dur="0.5"/> let me <pause dur="0.5"/> so i thought this would be a non-paper course it's just all going to be an electronic course <pause dur="0.3"/> and then i thought well you can't not give handouts <pause dur="0.4"/> and when <pause dur="0.4"/> when people come round universities doing teaching quality assessments <pause dur="0.4"/> all they do is they # they don't look at films like this <pause dur="0.4"/><kinesic desc="indicates camera" iterated="n"/> they # <pause dur="0.3"/> all they do <pause dur="0.2"/> is they # <pause dur="1.0"/> they go to filing cabinets and they flick through filing cabinets and they want to see lots of handouts <pause dur="0.2"/> okay so i've had to do a handout </u><gap reason="break in recording" extent="uncertain"/> <u who="nm0785" trans="pause"> now to some extent <pause dur="0.5"/> because i've just downloaded it all you know you <trunc>n</trunc> you now need not go and look at the home

page 'cause you've got it all here <pause dur="0.5"/> but what i'm going to be <trunc>tr</trunc> what i'm going to be doing as i go through the course is <pause dur="0.3"/> sometimes i'll be saying <pause dur="0.2"/> i'm not going to cover this in the lectures <pause dur="0.5"/> it's and it might not be fully explained in the in the download of it because i've updated the home page <pause dur="0.3"/> so periodically i will be adding things to the and and updating the home page <pause dur="0.2"/> when i do that i will try and put within the home page <pause dur="0.3"/> # a recent edition blinking <pause dur="0.2"/> # icon <pause dur="0.3"/> that <trunc>s</trunc> that <pause dur="0.2"/> that warns you that there is something new that you might want to look at <pause dur="0.2"/> so it might be something that you want to keep up to date with to some extent <pause dur="0.4"/> and indeed <pause dur="0.2"/> this relates <pause dur="0.6"/> what i've just given you now <pause dur="0.2"/> relates to this first topic <pause dur="0.2"/> i'll say a few words in a minute about the course outline i'm going to go through in these five weeks <pause dur="0.5"/> but this relates just to this topic <pause dur="0.7"/> # <pause dur="0.9"/> over Christmas i spent a good fortnight of my time putting all this together on the home page which is a slightly tedious

business <pause dur="0.5"/> # <pause dur="0.3"/> i haven't yet got round to putting all the material in for topic two <pause dur="0.6"/> but i will be doing that hopefully before next week <pause dur="0.9"/> # <pause dur="0.7"/> and for topics three four and five <pause dur="0.3"/> it <trunc>m</trunc> <pause dur="0.6"/> # as my time becomes more and more precious through term rather than through <pause dur="0.2"/> Christmas and New Year et cetera <pause dur="0.4"/> my ability to deliver on my promise to have everything on the home page <pause dur="0.3"/> might become less and less credible <pause dur="0.6"/> so # by <pause dur="0.3"/> topics three four and five finishing off the course <pause dur="0.3"/> it might simply say <pause dur="0.4"/> here's a link to a few of the things <pause dur="0.3"/> # but it's <trunc>n</trunc> might not be as complete as this but i intend it <trunc>th</trunc> i intend that it should be <pause dur="1.7"/> okay <pause dur="0.2"/> so that's all by way of <pause dur="0.2"/> of introduction <pause dur="0.6"/> # <pause dur="1.7"/> so let me turn now <trunc>m</trunc> to more somewhat more substantive matters albeit still by way of some introduction <pause dur="0.9"/> if you turn <pause dur="0.3"/> <trunc>d</trunc> in this handout to the third page of it <pause dur="0.9"/> that gives you the course outline <pause dur="1.5"/> and <pause dur="1.2"/> i've indicated here that we're going to be having five lecture topics <pause dur="0.3"/> now they might not correspond exactly <pause dur="0.5"/> because of my inability to time

things perfectly <pause dur="0.6"/> to <pause dur="0.3"/> each of our five meetings <pause dur="0.2"/> in fact some of these topics <trunc>in</trunc> inherently are bigger than others <pause dur="0.3"/> so for example <pause dur="0.4"/> this lecture one material that i'm going to <pause dur="0.3"/> to motivate in a few minutes <pause dur="0.4"/> # is going to certainly run into next week <pause dur="0.7"/> and i'd suspect that it will probably run into the <trunc>fu</trunc> first full hour of next week <pause dur="0.8"/> trade and wages will then occupy the second part and maybe just kick into week three <pause dur="0.4"/> and as we go through the topics become <pause dur="0.3"/> a bit smaller <pause dur="0.5"/> so we'll catch up with ourselves hopefully and cover all this material <pause dur="0.6"/> the first four topics <pause dur="0.3"/> are all pretty much integrated around the central theme <pause dur="0.6"/> and the theme is <pause dur="0.7"/> what determines wages in labour markets <pause dur="0.5"/> especially labour markets where there was imperfect competition <pause dur="0.3"/> both on the labour market side in terms of the presence of trade unions or other forms of bargaining by workers <pause dur="0.4"/> and also imperfect competition on the product market side <pause dur="1.0"/> now that's a <trunc>s</trunc> a theme which has <pause dur="0.5"/> still <pause dur="0.2"/> has relatively

little attention in the literature <pause dur="0.9"/> i i'll say a few more words about that in a minute <pause dur="0.8"/> so that's that's the main theme in a sense <pause dur="0.3"/> within that theme we say <pause dur="0.3"/> well this imperfect competition in the product market <pause dur="0.4"/> has an international <pause dur="0.4"/> competition element as well <pause dur="0.2"/> so product markets are typically <pause dur="0.3"/> internationally <pause dur="0.4"/> competed increasingly through globalization regionalizing <pause dur="0.4"/> of <trunc>econ</trunc> of of product markets <pause dur="0.4"/> # <pause dur="0.3"/> and through <trunc>int</trunc> in a European context through the increasing <trunc>ec</trunc> economic integration that we're seeing <pause dur="0.3"/> in a global sense but <trunc>al</trunc> but # specifically not least after the # the Euro was introduced <pause dur="0.4"/> in a <trunc>i</trunc> in a European context <pause dur="0.4"/> and i want to address questions to do with <pause dur="0.3"/> how that <pause dur="0.6"/> international context of product markets <pause dur="0.2"/> especially that increasing integration of product markets in the international context <pause dur="0.3"/> how that impacts on <pause dur="0.3"/> labour markets <pause dur="0.4"/> # wage determination in particular <pause dur="0.5"/> and of course the endogeneity of that <pause dur="0.3"/> how those <pause dur="0.3"/> labour market institutions and labour

market processes <pause dur="0.3"/> themselves <pause dur="0.2"/> shape the kind of influences that we'll expect to see from economic integration <pause dur="1.3"/> so much of that focus will be on a in a in a U-K European context <pause dur="0.4"/> but bearing in mind that there are analogies for <trunc>f</trunc> for North American labour markets <pause dur="0.4"/> # which we'll also be bearing in mind <pause dur="1.1"/> another aspect of that trade and wages literature of course is that that we we know that in the U-K <pause dur="0.4"/> and in the North America there have been tremendous growth in wage inequalities <pause dur="0.4"/> in in in the nineties perhaps starting earlier than that <pause dur="0.4"/> which in the wider European context <pause dur="0.2"/> has not seen <pause dur="0.4"/> the same impact <pause dur="0.4"/> # on wage inequality but there have been other impacts in terms of unemployment in the way that labour markets have adjusted <pause dur="0.5"/> and we want to address that question of <pause dur="0.3"/> to what extent have these changes in the labour market outcomes <pause dur="0.2"/> wage inequality and unemployment <pause dur="0.4"/> to what extent <pause dur="0.5"/> do those reflect <pause dur="0.3"/> the changing nature of trade that we're seeing in product markets or that they have

other causes <pause dur="0.4"/> and that will be our <pause dur="0.2"/> our second topic <pause dur="1.6"/> the third <pause dur="0.4"/> <trunc>a</trunc> and we'll be mostly focusing now on things to do with empirical evidence <pause dur="0.9"/> in the third topic <pause dur="0.4"/> i'm going to say <pause dur="0.5"/> # <pause dur="0.2"/> let's look at some recent models <pause dur="0.2"/> of <pause dur="0.8"/> the economic theory of trade unions in the context of international competition <pause dur="0.2"/> but in which we also look explicitly at international trade in the theoretical models <pause dur="0.9"/> # <pause dur="1.5"/> lecture four <pause dur="0.5"/> then still has that same theme of thinking about <pause dur="0.5"/> # <pause dur="0.5"/> # wages and institutions like unions and and like imperfect competition <pause dur="0.5"/> and more explicitly says well <pause dur="0.4"/> are these things of unions are these institutions are they exogenously determined do we just say some labour markets are unionized <pause dur="0.2"/> let's think about the impact of that <pause dur="0.4"/> or shouldn't we have <pause dur="0.4"/> more of a <trunc>r</trunc> of a role in saying well what determines whether we'll <pause dur="0.2"/> observe <pause dur="0.5"/> different kinds of bargaining arrangements in different labour markets <pause dur="0.3"/> so we look there at union formation <pause dur="0.7"/> # <pause dur="0.2"/> in a <trunc>eu</trunc> in a U-K context <pause dur="0.7"/> we ought to look at the

question of <pause dur="0.3"/> well <pause dur="0.4"/> this formation of unions <pause dur="0.9"/> this issue of whether unions are present or not <pause dur="0.3"/> isn't only to do with <pause dur="0.4"/> a supply side response by workers wanting to form unions <pause dur="0.2"/> but it might also depend on the attitudes of firms <pause dur="0.3"/> in whether they're prepared to accommodate unions <pause dur="0.7"/> and and in the U-K context that means looking at the issue of recognition <pause dur="0.4"/> the issue of whether firms <pause dur="0.2"/> recognize <pause dur="0.3"/> unions for the purposes of bargaining with them <pause dur="0.3"/> in the determination of wages and or other conditions <pause dur="0.8"/> that's very relevant as we as <trunc>w</trunc> at the moment because <pause dur="0.6"/> the new Labour government <pause dur="0.3"/> is proposing big changes in the way in which <pause dur="0.7"/> unions <pause dur="0.2"/> # <pause dur="0.3"/> can <pause dur="0.4"/> obtain recognition <pause dur="0.8"/> for the purposes of bargaining <pause dur="1.0"/> # <pause dur="0.8"/> in the <trunc>U</trunc> in the U-S for a long time there's been a long literature now <pause dur="0.4"/> on the fact that <pause dur="0.3"/> workers <pause dur="0.9"/> in the <trunc>f</trunc> # through organizing elections <pause dur="0.6"/> can # <pause dur="0.3"/> can have a legal right to have recognition by their by their employers <pause dur="0.6"/> in the U-K there haven't been there hasn't been that kind of legislative

environment and that's what's being introduced <pause dur="0.3"/> so we'll look at the context for that in terms of <pause dur="0.3"/> the <trunc>li</trunc> # any literature on recognition <pause dur="0.6"/> and a related issue to that is <pause dur="0.3"/> what determines why individual members might want to join unions <pause dur="0.8"/> # <pause dur="0.6"/> to the extent that unions are a public good <pause dur="0.5"/> or or a collective good <pause dur="0.9"/> then there's a there's a dilemma for <pause dur="0.2"/> for <pause dur="0.2"/> economic theory to explain why unions form <pause dur="0.5"/> let let me make that a bit more concrete <pause dur="1.2"/> suppose we're we we are a workforce <pause dur="0.3"/> the the eight or nine people of us in this room are a workforce <pause dur="0.5"/> # <pause dur="0.3"/> then we can all agree that jointly we might want to get together <pause dur="0.4"/> and form some <pause dur="0.2"/> some kind of bargaining unit let's call it a union might not always be formalized through channels like unions <pause dur="1.6"/> in the belief that <pause dur="0.2"/> if all of us in this workforce <pause dur="0.5"/> are are a <trunc>m</trunc> are a part of this group <pause dur="0.4"/> then our collective bargaining power is greater than our individual bargaining powers depending upon the nature of the work we do <pause dur="0.2"/> perhaps <pause dur="0.7"/> but if that is

the case <pause dur="0.5"/> then <pause dur="0.2"/> we can hope that by exerting that kind of bargaining power we can raise our wage <pause dur="0.5"/> if we're able to do that typically <pause dur="0.2"/> in labour markets <pause dur="0.3"/> any wage increase that we're able to bargain <pause dur="0.2"/> is then <trunc>bai</trunc> paid to all the members of that workforce <pause dur="0.7"/> not only to those people who have got themselves together <pause dur="0.2"/> # and and obtained through bargaining that wage <pause dur="1.1"/> so that generates a free rider problem for each and every one of us each one of us thinks well i'll let the other guys get together <pause dur="0.3"/> and spend their evenings and and their resources on forming this union <pause dur="0.2"/> and maybe also risking <pause dur="0.5"/> the wrath of employers depending on the attitude of of firms to unions <pause dur="1.1"/> and if they're successful in raising the wage <pause dur="0.4"/> i will then have a higher wage as well without incurring the costs <pause dur="0.4"/> so there's <pause dur="0.4"/> especially when there's a large workforce <pause dur="0.5"/> not when there's a small group like this perhaps but especially when there's a large workforce <pause dur="0.3"/> that kind of <pause dur="0.3"/> free rider incentive <pause dur="0.5"/> that kind of <pause dur="0.3"/>

prisoner's dilemma problem <pause dur="0.5"/> can undermine <pause dur="0.4"/> the generation of <pause dur="0.5"/> <trunc>u</trunc> # of unions of bargaining groups <pause dur="0.4"/> and so we'll be looking at questions <pause dur="0.2"/> to the <trunc>ex</trunc> <pause dur="0.2"/> # about have <trunc>y</trunc> have <pause dur="1.2"/> economists in the literature <pause dur="0.5"/> adequately explained that kind of thing <pause dur="0.8"/> and that's particularly important in a context in which <pause dur="0.3"/> union membership <pause dur="0.3"/> in economies like the U-K and in Europe generally <pause dur="0.5"/> # <pause dur="0.2"/> have observed <pause dur="0.3"/> such <pause dur="0.3"/> dramatic changes <pause dur="0.3"/> in the in the kind of levels of membership <pause dur="0.9"/> U-K <pause dur="0.4"/> is not necessarily representative throughout Europe but in the U-K <pause dur="0.8"/> in nineteen-seventy-nine something like fifty-five per cent of all workers were members of trade unions <pause dur="0.6"/> anyone know for the figure for nineteen-ninety <pause dur="0.2"/> where are we now nineteen-ninety-nine <pause dur="1.1"/> roughly </u><pause dur="1.2"/> <u who="sm0786" trans="pause"> thirty </u><pause dur="0.5"/> <u who="nm0785" trans="pause"> yeah it's possibly even less than thirty per cent around twenty-five twenty-seven per cent <pause dur="0.3"/> that's a dramatic change <pause dur="0.4"/> that's happened in a number of European countries it's not happening in all European countries <pause dur="0.3"/> indeed there's a literature on the

diversity <pause dur="0.2"/> and the causes of the diversity <pause dur="0.2"/> in union membership <pause dur="0.2"/> across European countries <pause dur="0.4"/> and not only the diversity in the level but the diversity in the pattern of change <pause dur="0.6"/> and that's something that we'll address in in topic four <pause dur="1.7"/> that then <pause dur="0.2"/> finishes the theme <pause dur="0.2"/> of <pause dur="0.9"/> of wages bargaining <pause dur="0.2"/> institutions imperfect competition <pause dur="0.5"/> # <pause dur="0.4"/> and in # and in some ways it it finishes the course <pause dur="0.2"/> and it might occupy all five of our meetings <pause dur="0.6"/> # <pause dur="0.4"/> but i'm <pause dur="0.3"/> wanting to have <pause dur="0.2"/> time to cover lecture five <pause dur="0.5"/> which if i find that our time is squeezed out by the other <pause dur="0.2"/> material <pause dur="0.3"/> becoming larger than than we <trunc>in</trunc> intend now <pause dur="0.4"/> i'll offer as a on a purely voluntary basis an additional lecture <pause dur="0.6"/> where i cover two other <pause dur="0.2"/> aspects <pause dur="0.5"/> # which are <pause dur="0.3"/> very relevant to current policies <pause dur="1.3"/> the first of those <pause dur="0.3"/> is <pause dur="0.2"/> what i call here <pause dur="0.3"/> # <pause dur="0.5"/> the analysis of monopsony <pause dur="1.1"/> # <pause dur="0.5"/> now the analysis of monopsony has A a very long tradition <pause dur="0.7"/> # and B a more recent tradition in terms of models of dynamic monopsony <pause dur="0.2"/> but i'm not concerned with either of those <pause dur="0.3"/>

i'm concerned with <pause dur="0.3"/> the issue of explaining <pause dur="0.3"/> not <pause dur="0.4"/> levels of employment and wages but <pause dur="0.3"/> the number of hours that people work <pause dur="0.9"/> at the moment <pause dur="0.6"/> in <pause dur="0.3"/> the European Union <pause dur="0.3"/> and impacting perhaps <pause dur="0.2"/> exclusively on the U-K <pause dur="0.4"/> you <pause dur="0.3"/> may well be aware that <pause dur="0.5"/> there is an E-U directive on maximum hours legislation <pause dur="0.7"/> which proposes <pause dur="0.4"/> a maximum of forty-eight hours per week <pause dur="0.4"/> # <pause dur="0.3"/> as a as a required contractual maximum by firms on workers <pause dur="0.3"/> firms and workers will be entitled <pause dur="0.4"/> to negotiate <pause dur="0.4"/> voluntarily and jointly <pause dur="0.3"/> hours in excess of forty-eight hours but firms will not be legally allowed <pause dur="0.2"/> to require workers <pause dur="0.2"/> without that kind of <pause dur="0.5"/> bargaining or voluntary contracting <pause dur="0.4"/> # it will they will not be require <pause dur="0.2"/> able to require workers to work <pause dur="0.3"/> for more than forty-eight hours per week <pause dur="0.6"/> now in many European countries that's irrelevant <pause dur="0.3"/> 'cause no one works that <pause dur="0.2"/> or very very few workers <pause dur="0.4"/> work that number of hours per week <pause dur="0.8"/> but the U-K is an outlier <pause dur="0.5"/> at least in Europe <pause dur="0.4"/> for having a very large number of workers <pause dur="0.2"/> who are working very long

hours <pause dur="0.6"/> and so i'm going to be concerned with <pause dur="0.4"/> addressing the question of <pause dur="0.4"/> why might workers be working very long hours <pause dur="0.2"/> and what might be the impacts of <pause dur="0.3"/> <trunc>br</trunc> introducing legislation on that <pause dur="0.3"/> and the context for examing that # examining those kind of issues will be <pause dur="0.3"/> # <pause dur="0.3"/> a comparison or a contrast of competitive models of labour markets with monopsonic models of labour markets <pause dur="0.2"/> in that context <pause dur="0.7"/> and <pause dur="0.9"/> and # underlining the <trunc>emph</trunc> <pause dur="0.2"/> the importance of that kind of issue <pause dur="0.6"/> in terms of that regulation of of <pause dur="0.3"/> <trunc>eur</trunc> of labour markets within Europe and impacting perhaps exclusively on the U-K <pause dur="0.3"/> at a time when other labour markets in Europe <pause dur="0.3"/> are bringing in national legislation <pause dur="0.4"/> to bring in maximum hours much less than that more like forty thirty-eight or even thirty-seven hours <pause dur="0.5"/> in the case of France <pause dur="0.3"/> Italy and Germany <pause dur="0.5"/> all through different methods <pause dur="0.2"/> some <pause dur="0.3"/> by being imposed by government some by being <pause dur="0.3"/> # voluntarily <pause dur="0.3"/> # negotiated through union and firm kind of arrangements <pause dur="1.2"/> # <pause dur="0.3"/> but at the same time as

that's happening in the U-K <pause dur="0.3"/> we also know that the new Labour government is bringing in minimum wage legislation which is another kind of regulation on the labour market <pause dur="0.3"/> i'm interested in <pause dur="0.3"/> the combination of those two forces <pause dur="0.4"/> on <pause dur="0.3"/> on labour market outcomes in terms of wages hours worked and employment <pause dur="1.0"/> okay <pause dur="0.3"/> so that's one of the <pause dur="0.5"/> # two issues i want to cover in lecture five <pause dur="0.7"/> and the second issue i want to look at <pause dur="0.3"/> is more to do with <pause dur="0.4"/> # <pause dur="0.4"/> # it's a <trunc>r</trunc> # a rather narrowly U-K focus <pause dur="0.4"/> and it's geared around some work that i'm doing with Jeremy Smith in the department <pause dur="0.4"/> # in which we're looking at <pause dur="0.4"/> evidence on U-K students <pause dur="0.4"/> and addressing questions like <pause dur="0.5"/> # does your does A your degree performance but <trunc>m</trunc> perhaps more importantly for our focus on this course <pause dur="0.4"/> do your earnings after graduation <pause dur="0.6"/> vary very much <pause dur="0.3"/> with the following kinds of characteristics <pause dur="0.2"/> firstly <pause dur="0.3"/> which university you went to <pause dur="0.3"/> so is there a big premium on going to some universities and not others <pause dur="0.5"/> secondly <pause dur="0.2"/> the subject you've taken <pause dur="0.8"/>

is there a big premium for economics versus business studies or whatever <pause dur="0.8"/> # <pause dur="0.5"/> and thirdly perhaps more interestingly <pause dur="0.3"/> to what extent <pause dur="0.2"/> do your does your labour market outcome <pause dur="0.4"/> depend upon your personal characteristics <pause dur="0.6"/> # <pause dur="0.4"/> age and gender being <pause dur="0.2"/> # marital status being one set <pause dur="0.3"/> but others being <pause dur="0.6"/> # <pause dur="0.4"/> your <pause dur="0.2"/> schooling schooling information <pause dur="0.3"/> prior to university so <pause dur="0.5"/> does it depend on what kind of school you went to <pause dur="0.7"/> and are there big differences for example between <pause dur="0.3"/> private schools and state schools in your labour market outcomes even for those people who graduate <pause dur="0.6"/> does it also depend upon the degree class you obtain <pause dur="0.7"/> # <pause dur="0.3"/> and does it depend upon demographic characteristics or social background characteristics <pause dur="0.4"/> such as the <pause dur="0.4"/> occupational background of the <pause dur="0.3"/> <trunc>f</trunc> <pause dur="0.2"/> of the parental background from which you <pause dur="0.2"/> you come <pause dur="0.6"/> so that's one set of issues <pause dur="0.2"/> related to that <pause dur="0.5"/> issue <pause dur="0.5"/> is <pause dur="0.4"/> # <pause dur="0.4"/> how can we explain different performances in the labour market across different universities <pause dur="0.2"/> if we observe a big premium for

going to Oxbridge rather than to other universities <pause dur="0.4"/> is that <pause dur="0.3"/> because of some kind of signalling in the labour market some kind of discrimination maybe in the labour market <pause dur="0.4"/> or is it to do with <pause dur="0.4"/> # <pause dur="0.2"/> different characteristics of those universities which can be explained <pause dur="0.3"/> which aren't some kind of residual <pause dur="0.3"/> to be attributed perhaps to discrimination <pause dur="0.8"/> okay <pause dur="0.3"/> so those are the things to do to do in lecture five <pause dur="1.1"/> okay <pause dur="0.5"/> so i've spent quite a bit of time motivating lectures two to four to five and had # <trunc>s</trunc> said very little about lecture one so now now let's get <pause dur="0.3"/> down to the <pause dur="0.4"/> business of looking at the material for lecture one <pause dur="1.8"/> so <pause dur="0.9"/> this <pause dur="0.4"/> material <pause dur="3.1"/><kinesic desc="puts on transparency" iterated="n"/> for lecture one and hopefully this is going to be a sufficiently large font size <pause dur="4.1"/> <trunc>o</trunc> our our heading our title is <pause dur="0.2"/> wages bargaining and product market competition <pause dur="0.7"/> with <gap reason="name" extent="1 word"/> <pause dur="0.2"/> in the first part of # <pause dur="0.6"/> the academic year in this on this course in part one <pause dur="0.4"/>

you have spent <pause dur="0.4"/> # <pause dur="0.2"/> one part of your course <pause dur="0.3"/> looking at wages and bargaining <pause dur="0.7"/> so i don't intend to replicate that kind of thing <pause dur="0.3"/> my focus will be more on <pause dur="0.3"/> saying well let's <pause dur="0.5"/> extend that analysis of wages and bargaining <pause dur="0.3"/> to a context in which <pause dur="0.5"/> product markets are not necessarily perfectly competitive <pause dur="0.3"/> nor necessarily monopolistic <pause dur="1.1"/> say a bit more about why i'm going to do that in a moment <pause dur="1.3"/><kinesic desc="reveals covered part of transparency" iterated="n"/> this <pause dur="0.6"/> <trunc>m</trunc> <trunc>m</trunc> material on lecture one is going to be split into two <pause dur="0.5"/> one-A looks at the theoretical motivation <pause dur="0.3"/> for <trunc>wh</trunc> why we might be interested in this <pause dur="0.3"/> theoretical issues <pause dur="0.4"/> and then part <pause dur="0.2"/> one-B <pause dur="0.3"/> we'll look at the empirical evidence on these kinds of issues that we're going to be motivating <pause dur="0.9"/> in this <pause dur="0.6"/> hour <pause dur="0.5"/> i'm going to be <pause dur="0.5"/> focusing exclusively on theoretical motivation <pause dur="0.3"/> sometime in the second hour i'll get round to empirical evidence <pause dur="0.3"/> is there a problem in terms of </u><pause dur="0.5"/> <u who="sm0787" trans="pause"> yes </u><u who="nm0785" trans="latching"> visibility or the handout </u><u who="sf0788" trans="latching"> no </u><u who="sm0787" trans="overlap"> no it's # </u><pause dur="0.3"/> <u who="sf0788" trans="pause"> it's different </u><u who="sm0787" trans="overlap">

it's a different <pause dur="0.3"/> it's a bit different <gap reason="inaudible" extent="1 sec"/> </u><u who="nm0785" trans="overlap"> it's a bit <pause dur="0.2"/> a little different from this <pause dur="1.0"/> is that right <pause dur="1.7"/><event desc="looks through notes" iterated="y" dur="2"/> # </u><u who="sf0789" trans="latching"> it is different </u><u who="nm0785" trans="latching"> i don't think it is we're doing lecture one product market influences </u><u who="sm0790" trans="overlap"> <gap reason="inaudible" extent="1 sec"/></u><u who="nm0785" trans="overlap"> # on wages <pause dur="0.2"/> okay </u><pause dur="0.5"/> <u who="sm0790" trans="pause"> oh okay <gap reason="inaudible" extent="1 sec"/> </u><u who="nm0785" trans="overlap"> different way of saying the same thing sure <pause dur="0.2"/> okay <pause dur="0.4"/> and then we're looking at the theoretical motivation <pause dur="0.2"/> sure <pause dur="2.1"/> # <pause dur="1.5"/> a general point <pause dur="0.7"/> we're a small group this must be interactive <pause dur="0.4"/> so do raise issues and that's why i was looking for the facial expressions okay i'm very keen to have this interactive and <trunc>respon</trunc> <pause dur="0.8"/> okay <pause dur="0.4"/> so <pause dur="0.8"/> raise an eyebrow at your peril 'cause i will ask you to say what the <unclear>question</unclear> is <vocal desc="laughter" iterated="y" dur="1"/> </u><pause dur="0.3"/> <u who="om0791" trans="pause"> may i raise a point </u><u who="nm0785" trans="latching"> yeah of course </u><u who="om0791" trans="latching"> # sorry is this weeks six to ten there is that </u><u who="nm0785" trans="latching"> yep <pause dur="0.3"/> # <pause dur="0.3"/> that's weeks to <trunc>s</trunc> # six to ten of this course </u><u who="om0791" trans="overlap"> ah </u><pause dur="0.3"/> <u who="nm0785" trans="pause"> # <pause dur="0.6"/> the students have already taken weeks

one to five </u><pause dur="0.2"/> <u who="om0791" trans="pause"> ah right </u><u who="nm0785" trans="overlap"> of the course sorry it doesn't relate to term time it's <vocal desc="laughter" iterated="y" n="om0791" dur="1"/> it's a good point it it's </u><u who="om0791" trans="overlap"> <unclear>no</unclear> </u><u who="nm0785" trans="overlap"> potentially very confusing <pause dur="0.5"/> # <pause dur="1.2"/> okay <pause dur="1.1"/> # so <pause dur="0.3"/> in this issue of theoretical motivation <pause dur="0.5"/><kinesic desc="reveals covered part of transparency" iterated="n"/> # <pause dur="0.6"/> i'm going i'm <trunc>ar</trunc> i'm arguing that <pause dur="0.2"/> that there are two questions which are going to concern us for this part of the course <pause dur="0.8"/> and one question is <pause dur="0.9"/> what generates rents <pause dur="0.4"/> if we're looking at what determines wages <pause dur="0.6"/> and we're saying we're not just going to <pause dur="0.3"/> assume that wages are <pause dur="0.2"/> given by competitive levels <pause dur="1.0"/> but they might differ from competitive levels <pause dur="0.5"/> then we <trunc>ha</trunc> have to explain <pause dur="0.4"/> why there might be rents <pause dur="0.3"/> which enable workers to have <pause dur="0.3"/> wages <pause dur="0.5"/> in excess of <pause dur="0.4"/> competitive levels <pause dur="0.6"/> so the question <trunc>i</trunc> <trunc>i</trunc> <trunc>w</trunc> <trunc>wh</trunc> a question which is very important is <pause dur="0.2"/> what generates rents <pause dur="0.7"/> with <gap reason="name" extent="1 word"/> in in the previous <pause dur="0.3"/> weeks of this course <pause dur="0.3"/> you'll have spent quite a bit of time looking at issues to do with rent sharing 'cause that's

something which <gap reason="name" extent="1 word"/>'s done a lot of work on <pause dur="0.5"/> and <pause dur="0.4"/> and one of his assumptions is that there are rents <pause dur="0.5"/> and then <pause dur="0.7"/> <trunc>a</trunc> and he <trunc>m</trunc> and he <pause dur="0.2"/> demonstrates using a lot of evidence <pause dur="0.4"/> # about the existence of these rents and the and speculates what are the determinants of those rents <pause dur="0.6"/> so he's looked a lot at rent sharing <pause dur="0.6"/> i'm <pause dur="0.7"/> going to be <pause dur="0.3"/> initially asking the question well what does what generates those rents <pause dur="0.3"/> but primarily for us then to address the question <pause dur="0.3"/> of <pause dur="0.7"/> assuming those rents exist <pause dur="0.3"/> <reading>under what conditions are workers best able to capture a share of those rents</reading> <pause dur="0.4"/> 'cause you can imagine situations in which firms are generating huge rents because of monopoly power <pause dur="0.7"/> but that same monopoly power <pause dur="1.7"/> or <pause dur="0.4"/> reasons <pause dur="0.2"/> associated with that monopoly power <pause dur="0.3"/> might give those firms a great bargaining power over workers and prevent those workers from <trunc>ex</trunc> <trunc>ec</trunc> <pause dur="0.2"/> from generating <pause dur="0.8"/> # a share of those rents <pause dur="0.2"/> from capturing a share

of those rents <pause dur="0.4"/> okay indeed it might be the case that firms which are very powerful in their product markets <pause dur="0.3"/> also exert enormous <trunc>m</trunc> <pause dur="0.4"/> labour market power <pause dur="0.3"/> and <trunc>com</trunc> <pause dur="0.2"/> and # which enable them to pay wages below competitive levels <pause dur="1.5"/> so <pause dur="0.5"/> it it doesn't follow that just because rents exist workers are going to be <pause dur="0.2"/> capturing a share of them <pause dur="0.5"/> so we'll be # addressing the question of what <trunc>ena</trunc> <reading>under what conditions are workers best able to capture a share <pause dur="0.3"/> of any such rents</reading> <pause dur="0.3"/> which firms are <kinesic desc="reveals covered part of transparency" iterated="n"/> are are are are are <trunc>re</trunc> are receiving <pause dur="1.4"/> so what are the <trunc>re</trunc> sources of rents on the first <pause dur="0.2"/> of those two questions what are the sources <pause dur="0.4"/> of any such rents <pause dur="1.6"/> the most <pause dur="0.2"/> obvious starting point is to say that <pause dur="0.7"/> imperfect competition in the product market can generate rents <pause dur="2.4"/> two <pause dur="0.6"/> forms of <pause dur="0.2"/> product market imperfect competition would be monopoly and duopoly <pause dur="0.9"/> models <pause dur="1.2"/> of unions the economic theory of the trade union <pause dur="1.1"/> almost exclusively <pause dur="0.2"/> focuses <pause dur="0.4"/> on monopoly <pause dur="0.9"/> okay <pause dur="0.4"/> you will have looked at a number of <pause dur="0.4"/> of

models <pause dur="0.6"/> where <pause dur="1.1"/> you <trunc>s</trunc> you say that <pause dur="0.7"/><kinesic desc="writes on board" iterated="y" dur="20"/> there's a union <pause dur="0.4"/> downward sloping labour demand curve <pause dur="1.7"/> wages employment <pause dur="0.3"/> and maybe unions have got <trunc>s</trunc> # in a monopoly in a model <pause dur="0.7"/> unions are trying to <pause dur="2.0"/> # <pause dur="0.3"/> aim at <pause dur="0.5"/> this utility maximizing point <pause dur="0.4"/> with a wage in excess of <pause dur="0.6"/> some competitive wage <pause dur="0.9"/> and so they're extracting some rents in the form of wages being higher than they otherwise would be <pause dur="0.3"/> and those rents are pretty much assumed to come from <pause dur="0.2"/> monopoly power <pause dur="0.6"/> by firms <pause dur="0.4"/> or <pause dur="0.5"/> if not from that <pause dur="0.3"/> from a second <pause dur="0.3"/><kinesic desc="reveals covered part of transparency" iterated="n"/> source <pause dur="0.4"/> of <trunc>p</trunc> of rents <pause dur="0.4"/> and this is arguably even more common in the literature <pause dur="1.3"/> this this source being <pause dur="0.3"/> the diminishing marginal <pause dur="0.5"/> product of labour <pause dur="0.5"/> okay so <trunc>i</trunc> <pause dur="0.2"/> in models like that <pause dur="1.0"/> you've <pause dur="0.7"/> the argument goes that the firm's profits <pause dur="0.7"/><kinesic desc="writes on board" iterated="y" dur="15"/> are equal to price <pause dur="0.2"/> which is <pause dur="1.8"/> parametric exogenously given <pause dur="0.8"/> at some competitive price level <pause dur="1.4"/> times output <pause dur="1.4"/> output being some function of labour inputs <pause dur="0.4"/> minus <pause dur="0.8"/> costs <pause dur="0.2"/>

themselves some function of wages and so often the argument goes that <pause dur="0.3"/> the <kinesic desc="writes on board" iterated="y" dur="1"/><kinesic desc="indicates point on board" iterated="n"/> price is parametric price is exogenously given at some competitive price level <pause dur="0.6"/> and then <pause dur="0.8"/> what's enabling workers to have <pause dur="0.3"/> wages above a competitive level <pause dur="0.3"/> is that <pause dur="1.7"/> the production function exhimits exhibits diminishing marginal product of labour <pause dur="5.0"/><kinesic desc="writes on board" iterated="y" dur="4"/> and <pause dur="0.4"/> that then is used <pause dur="0.3"/> to # <pause dur="0.2"/> to generate the result that wages <pause dur="0.6"/> set by unions or bargained between unions and firms <pause dur="0.3"/> might be greater than the competitive level <pause dur="0.6"/> now <pause dur="0.5"/> if you think about <kinesic desc="indicates board" iterated="n"/> that model <pause dur="0.4"/> it's not <pause dur="0.4"/> very satisfying from an empirical point of view <pause dur="0.5"/> because <pause dur="0.3"/> if the <pause dur="0.3"/> product market is <pause dur="0.3"/> competitive <pause dur="0.7"/> then we're going to assume pretty much that at least <pause dur="0.9"/> even in the medium term but certainly in the long run <pause dur="0.3"/> firms are not going to be able to persist in paying <pause dur="0.3"/> wages above competitive levels <pause dur="0.6"/> okay <pause dur="0.4"/> so a much more satisfactory it seems to me <pause dur="0.9"/> motivation for the existence of rents being used to share <pause dur="0.3"/> is the existence of <pause dur="0.3"/>

imperfect competition in the product market <pause dur="2.2"/> and <pause dur="0.3"/> monopoly is an extreme version of that which the literature <pause dur="0.4"/> has <pause dur="0.2"/> taken on board and has focused on when it's not <pause dur="0.4"/> adopted this kind of approach <pause dur="0.6"/> but of course we know that monopoly is <pause dur="0.5"/> a rather extreme form of market competition perhaps no more likely than perfect competition <pause dur="0.8"/> and <pause dur="0.3"/> it strikes me that oligopoly <pause dur="1.0"/> # i have <trunc>in</trunc> indicated here two firm oligopoly but perhaps <trunc>m</trunc> i should <trunc>s</trunc> more generally say oligopoly <pause dur="0.4"/> is a more likely context to explain the existence of rents being as we capture them <pause dur="0.4"/> so what i'm going to be wanting to do quite early on in <pause dur="0.2"/> <trunc>i</trunc> in this hour <pause dur="0.4"/> is look <pause dur="0.3"/> at <pause dur="0.3"/> extending the traditional model of union-<pause dur="0.2"/>firm bargaining <pause dur="0.3"/> to a context where there's oligopoly <pause dur="0.6"/> and <pause dur="0.4"/> and a reason for doing that is because that forces us <pause dur="0.4"/> to look to think more seriously and more formally about interaction between labour markets and product markets <pause dur="0.3"/> which <kinesic desc="indicates board" iterated="n"/> these kind of models ignore <pause dur="0.7"/> because <pause dur="1.1"/> if <pause dur="0.4"/> we're dealing either with perfect competition where price is

parametric <pause dur="0.3"/> or with monopoly <pause dur="0.2"/> unless we're thinking about entry deterrents <pause dur="0.3"/> then there there's no strategic behaviour going on in the product market <pause dur="0.6"/> and perfect competition <pause dur="0.5"/> there's no strategic behaviour because you are a such a small part of the big product market you needn't think about the impact of your behaviour <pause dur="0.2"/> on other firms <pause dur="0.6"/> and if you're a monopolist <pause dur="0.2"/> there are no other firms so again there's no strategic behaviour <pause dur="0.5"/> so those two traditional approaches <pause dur="2.1"/> free us from the requirement to <pause dur="0.7"/> or more critically don't <pause dur="0.9"/> encourage us to think <pause dur="0.3"/> more formally about <pause dur="0.2"/> the importance <pause dur="0.2"/> what will be in the real world the important interactions between <pause dur="0.8"/> labour markets and product market outcomes in terms of strategic behaviour <pause dur="0.5"/> let me give you a more concrete example on all of that <pause dur="0.4"/> if we are a workforce <pause dur="0.5"/> if we are a unionized workforce <pause dur="0.2"/> and we are <pause dur="0.2"/> determining what wage to set <pause dur="0.3"/> do we merely think in terms of the wage and employment combinations <pause dur="0.6"/> in some <pause dur="0.2"/> way <pause dur="0.3"/> abstracted from <pause dur="0.6"/> the the

wider world outside us or do we recognize that the higher is our wage <pause dur="1.2"/> the less <pause dur="0.2"/> potentially at least <pause dur="1.0"/> the the the # <pause dur="0.2"/> potentially at least <pause dur="0.4"/> the higher is our wage <pause dur="0.4"/> the less competitive will be the product market position of our firm in its <pause dur="0.5"/> product market <pause dur="0.2"/> decision making <pause dur="0.5"/> so we <pause dur="0.6"/> perhaps ought to be aware as a union of the dangers of setting a high wage <pause dur="0.9"/> # in terms of the impact that that will have on the product market competition <pause dur="0.5"/> # competitiveness <pause dur="0.5"/> of the firm that we're bargaining with <pause dur="0.7"/> # <pause dur="0.9"/> and so there are strategic behaviours that we should consider and that's what we'll analyse in a few moments <pause dur="0.7"/><kinesic desc="reveals covered part of transparency" iterated="n"/> # <pause dur="0.4"/> course other <trunc>f</trunc> <pause dur="0.5"/> other forms <pause dur="0.5"/> # <pause dur="0.2"/> of the generation of rents <pause dur="0.3"/> would include imperfect competition in the labour market rather than the product market <pause dur="0.8"/> okay so <pause dur="0.4"/> it might <pause dur="0.8"/> be for example that firms have monopoly power which is <trunc>generat</trunc> monopsony power which is generating rents for them <pause dur="3.1"/> and there is an increasing literature on <pause dur="0.5"/><kinesic desc="reveals covered part of transparency" iterated="n"/> monopsony power <pause dur="3.4"/> or other <trunc>v</trunc>

other forms of imperfect competition in the labour market might <pause dur="0.3"/> you might interpret efficiency wage models and even insider-outsider models in terms of <pause dur="0.3"/> imperfectly competitive labour markets <pause dur="0.8"/> okay <pause dur="0.5"/> and see those as forms of of rent generation <pause dur="0.4"/> but i'm going to pretty much exclusively concentrate <pause dur="0.4"/> on this issue of oligopoly <pause dur="0.6"/> so what i want to do <pause dur="0.6"/> # <kinesic desc="reveals covered part of transparency" iterated="n"/> within this theoretical motivation <pause dur="0.5"/> is look at actually three models <pause dur="0.7"/> of <pause dur="0.6"/> union-firm bargaining in a wider context of imperfect competition in the product market <pause dur="0.3"/> in order for us to get some feel for the kind of important issues that are <pause dur="0.4"/> that are going on there <pause dur="1.5"/><kinesic desc="reveals covered part of transparency" iterated="n"/> so let's look at the three models the first of and i label the models one-one-one <pause dur="0.3"/> one-one-two and one-one-three <pause dur="0.7"/> # <pause dur="0.3"/> and the first two of these <pause dur="0.6"/> are closely related to a paper published by Steve Dowrick in the Economic Journal <pause dur="0.5"/> of nineteen-eighty-nine <pause dur="5.7"/><event desc="takes off transparency" iterated="n"/> okay so let's think first <pause dur="0.2"/> about <pause dur="0.7"/> # <trunc>th</trunc> the model one <pause dur="2.8"/> and model one <pause dur="0.4"/> is <pause dur="0.3"/>

sketched <pause dur="0.7"/> on the <pause dur="0.2"/> home page <pause dur="0.4"/> a copy of which you've got <pause dur="0.5"/> # if you click on your handouts <pause dur="0.4"/> which means in this context <pause dur="0.4"/> scanning just a few pages <pause dur="0.5"/> you'll find something called model one-<pause dur="0.6"/>point-one-<pause dur="0.4"/>point-one <pause dur="0.9"/> and i will just # go through the key points of that <pause dur="0.4"/> on the board <pause dur="1.4"/><event desc="looks for board rubber" iterated="y" dur="uncertain"/> and it will be easier once i discover where the board rubber is </u><gap reason="break in recording" extent="uncertain"/> <u who="nm0785" trans="pause"> so thinking about this first model <pause dur="1.5"/> okay we're going to # make a number of simplifying assumptions <pause dur="0.4"/> the first one <pause dur="0.3"/> will be that <pause dur="0.6"/> product demand is linear <pause dur="0.6"/> so you'll <pause dur="0.5"/> # <pause dur="1.0"/> so we'll write <pause dur="0.7"/><kinesic desc="writes on board" iterated="y" dur="3"/> product demand <pause dur="1.5"/> of some linear <pause dur="0.9"/> # <pause dur="1.6"/> indirect demand function <pause dur="1.4"/> where <pause dur="1.1"/><kinesic desc="writes on board" iterated="y" dur="1"/> X is market outputs <pause dur="0.6"/> of a homogeneous <pause dur="0.2"/> commodity <pause dur="1.0"/> in model two we'll look at <pause dur="0.4"/> differentiated commodities and see that that <pause dur="0.2"/> adds something to our analysis informs our analysis <pause dur="0.4"/> but for now assume a homogeneous commodity <pause dur="0.4"/> X <pause dur="1.0"/> # <pause dur="0.9"/> is the quantity of that <pause dur="0.5"/> and that is being supplied by N firms we're dealing initially with an N

firm <pause dur="0.3"/> oligopoly <pause dur="1.7"/><kinesic desc="writes on board" iterated="y" dur="9"/> so the market output X is the sum of the outputs of all the individual firms which we'll <pause dur="1.1"/> label <pause dur="0.4"/> I <pause dur="0.3"/> I going from one to N <pause dur="2.7"/> we'll assume a constant marginal product of labour <pause dur="1.1"/> and we'll take that as our numeraire <pause dur="1.3"/> in other words marginal product of labour is equal to one <pause dur="0.8"/> the reason for doing that <pause dur="0.4"/> it simplifies things very nicely <pause dur="0.2"/> we can just <pause dur="0.4"/> label outputs <pause dur="0.2"/> and employment as the same thing it means output is <trunc>e</trunc> equals employment <pause dur="1.3"/> one unit one unit of labour generates one unit of output <pause dur="0.7"/> we could <pause dur="0.5"/> easily <pause dur="0.6"/> <trunc>ec</trunc> generate out from that it wouldn't it doesn't change things too much so long as we keep the assumption of constant marginal product of labour <pause dur="1.3"/> we're going to assume a two stage game <pause dur="0.6"/> in which <pause dur="0.2"/><kinesic desc="writes on board" iterated="y" dur="10"/> in stage one <pause dur="3.7"/> each union <pause dur="1.9"/> and firm <pause dur="1.3"/> pair <pause dur="0.9"/> 'cause we assume that each one of these N firms <pause dur="0.5"/> is <trunc>bar</trunc> <pause dur="0.2"/> <trunc>i</trunc> <trunc>i</trunc> confronts <pause dur="0.4"/> a local union <pause dur="0.7"/> so we're going to have decentralized bargaining between each union and firm <pause dur="0.6"/> pair we'll

call it a union-firm bargaining pair <pause dur="0.5"/> so each union-firm <pause dur="0.8"/><kinesic desc="writes on board" iterated="y" dur="8"/> bargaining pair <pause dur="1.9"/> sets the wage <pause dur="1.9"/> in that <pause dur="1.8"/> in that for that pair so <pause dur="0.2"/> W-I for <trunc>f</trunc> <pause dur="0.3"/> <trunc>p</trunc> pair I <pause dur="1.6"/> as a <trunc>resul</trunc> as a <pause dur="0.7"/> as a <pause dur="0.2"/><kinesic desc="writes on board" iterated="y" dur="2"/> result of bargaining <pause dur="0.5"/> bargaining being described <pause dur="0.4"/> in the <pause dur="0.8"/> in a traditional right to manage model <pause dur="1.1"/> okay so that there there's bargaining going on over the wage <pause dur="0.2"/> but the firm is then <pause dur="0.3"/> going to be <pause dur="0.5"/> at some point it'll actually be in stage two <pause dur="0.3"/> is going to be setting employment <pause dur="0.4"/> autonomously <pause dur="0.3"/> in other words the union has no influence over employment <pause dur="0.9"/> in this model <pause dur="1.9"/> when each union-firm pair sets <pause dur="0.4"/> its wage W-I <pause dur="0.3"/> it <pause dur="0.3"/> it <pause dur="1.1"/> takes as given in a kind of Cournot way <pause dur="0.4"/> Cournot-Nash way <pause dur="0.2"/> it takes as given <pause dur="0.3"/> any wage set <pause dur="0.4"/> in any other union-firm bargaining pair <pause dur="0.6"/> so it doesn't <pause dur="1.2"/> take on board in a kind <trunc>i</trunc> in # <pause dur="1.2"/> in any kind of Stackelberg way <pause dur="0.5"/> the # <pause dur="0.2"/> the choice being made by others the impact of its choice on others <pause dur="0.4"/> okay <pause dur="1.4"/> in <pause dur="0.5"/> in # <pause dur="0.8"/> we we can <trunc>exten</trunc> we can easily extend that in fact i might as i'm doing this

extend that <pause dur="1.4"/> # <pause dur="1.6"/> okay so let's let me write that down in that way <pause dur="0.5"/> if each firm is spending <pause dur="0.3"/> let's let's <pause dur="1.0"/><kinesic desc="writes on board" iterated="y" dur="6"/> say <pause dur="0.2"/> that more generally <pause dur="0.4"/> actually <pause dur="0.7"/> the Cournot-Nash <pause dur="0.4"/> assumption of course is a <trunc>w</trunc> a specific form where you assume that <pause dur="0.6"/> your action <pause dur="0.2"/> doesn't affect the actions of others <pause dur="0.3"/> you take the actions of others as given <pause dur="0.9"/> more generally we could use a conjectural variation approach <pause dur="0.2"/> which many people don't like and if you don't then just set the <trunc>l</trunc> conjectural variation parameter <pause dur="0.2"/> equal to <pause dur="0.2"/> zero <pause dur="0.3"/> and assume you're dealing with a Cournot model <pause dur="0.7"/> the <trunc>con</trunc> a conjectural <kinesic desc="writes on board" iterated="y" dur="3"/> variation <trunc>pr</trunc> <pause dur="0.5"/> approach <pause dur="0.8"/> would say that <pause dur="0.4"/> when we're setting <pause dur="0.2"/> as a union-firm pair <pause dur="0.3"/> when we're setting our wage <pause dur="1.7"/> # <pause dur="0.5"/> we allow for the possibility that <pause dur="0.3"/> the choice of our wage influences the wage choice of the other union-firm pairs <pause dur="0.6"/> according to some conjecture <pause dur="0.2"/> captured by the conjectural variation <pause dur="0.4"/> if that parameter is zero it means you assume that your <pause dur="0.6"/> wage doesn't affect the others <pause dur="0.3"/>

and that's like adopting <pause dur="0.3"/> a kind of best reply approach a Cournot approach <pause dur="0.7"/> where you just <pause dur="0.3"/> reply to if there's a choice in yours <pause dur="0.3"/> not taking into account how your wage impacts on the others </u><pause dur="0.5"/> <u who="sm0792" trans="pause"> in the model which this kind of # <unclear>game</unclear> result <pause dur="0.2"/> just in the symmetries <pause dur="0.5"/> in in which <pause dur="0.2"/> # <pause dur="1.0"/> the two <pause dur="0.8"/> different <pause dur="0.2"/> are <pause dur="0.8"/> way </u><pause dur="0.5"/> <u who="nm0785" trans="pause"> okay <pause dur="0.5"/> what i'm <pause dur="0.5"/> going to be assuming <pause dur="0.4"/> for these purposes <pause dur="0.4"/> # <pause dur="0.3"/> is that <pause dur="0.3"/> all of these union-firm pairs are identical <pause dur="0.4"/> that's is that the part of the point you're making that we that they're <trunc>s</trunc> that <pause dur="0.7"/> so the equilibrium we're going to generate <pause dur="0.2"/> will be a symmetric equilibrium <pause dur="1.3"/> and indeed at one point in the analysis we're going to use that <pause dur="0.5"/> to simplify the the solution to the problem <pause dur="0.9"/> but we're not going to be imposing it on the problem from the beginning </u><pause dur="0.2"/> <u who="sm0792" trans="pause"> <gap reason="inaudible" extent="1 sec"/></u><pause dur="0.2"/> <u who="nm0785" trans="pause"> we <pause dur="0.2"/> # # in # it's a good point i'll bear it in mind as i'm going through the model to try and explain what i mean by that <pause dur="0.4"/> # <pause dur="0.5"/> so the result will be <pause dur="0.7"/> one in which <pause dur="0.2"/> # we get symmetry but we can't impose right from the beginning <pause dur="0.3"/> the knowledge that in a symmetric

equilibrium <pause dur="0.2"/> wage will be the same <pause dur="1.0"/> just as when you're looking at <pause dur="0.4"/> # a simple Cournot product market story <pause dur="0.5"/> you know that in equilibrium <pause dur="0.3"/> in under symmetry <pause dur="0.4"/> # the the output choices of the two firms will be the same <pause dur="0.6"/> but you don't impose on the beginning you let that come out of an # of an equilibrium solution to the model <pause dur="0.2"/> and that's what we'll be doing here <pause dur="1.0"/> okay <pause dur="0.7"/> so we could have a <pause dur="0.3"/> a a general <trunc>convectj</trunc> <pause dur="0.2"/><kinesic desc="writes on board" iterated="y" dur="1"/><kinesic desc="indicates point on board" iterated="n"/> conjectural variation approach and indeed in model one-one-one that's that's what we're going to do <pause dur="1.1"/> okay so that's what's that's how the wages are being determined <pause dur="0.8"/> once the wage has been determined <pause dur="0.4"/> we then <kinesic desc="writes on board" iterated="y" dur="10"/> move into stage two <pause dur="3.0"/> in which <pause dur="0.5"/> with wages <pause dur="0.5"/> we could even say with the full vector of wages <pause dur="0.4"/> 'cause of course <kinesic desc="indicates point on board" iterated="n"/> this is generating a vector of wages <pause dur="0.3"/> W-I <pause dur="0.6"/> I going through one to N <pause dur="1.0"/> the vector of <pause dur="0.2"/> of wages <pause dur="1.5"/> # <pause dur="1.3"/> will be determined <kinesic desc="writes on board" iterated="y" dur="14"/> from <pause dur="0.4"/> stage one <pause dur="1.3"/> so it's been predetermined <pause dur="1.4"/> and firms then <pause dur="1.5"/> choose <pause dur="1.5"/> output <pause dur="0.5"/> and of course that's

the same as choosing <pause dur="0.7"/> employment <pause dur="0.3"/> on this assumption of constant marginal product of labour <pause dur="0.6"/> especially with our numeraire assumption <pause dur="3.2"/> firms <pause dur="0.4"/> how do firms choose output <pause dur="0.2"/> well we're going <pause dur="0.6"/> to <kinesic desc="writes on board" iterated="y" dur="7"/> <pause dur="0.3"/> adopt <pause dur="0.2"/> again the conjectural variation approach <pause dur="2.0"/> which just generalizes <pause dur="0.7"/> the Cournot-Nash approach <pause dur="0.5"/> of <pause dur="0.2"/> of <pause dur="0.2"/> generating best reply functions and again if you don't like <kinesic desc="indicates point on board" iterated="n"/> this idea <pause dur="0.3"/> just assume that the conjectural variation of it is zero <pause dur="0.3"/> # and look at the think about generating the Cournot-Nash equilibrium <kinesic desc="writes on board" iterated="y" dur="1"/><kinesic desc="indicates point on board" iterated="n"/> </u><pause dur="1.7"/> <u who="sm0793" trans="pause"> do you mean <gap reason="inaudible" extent="1 sec"/> <pause dur="0.7"/> firms will either choose to compete on price or <gap reason="inaudible" extent="1 sec"/> </u><u who="nm0785" trans="overlap"> we're going to assume the competition is on quantity <pause dur="0.6"/> okay <pause dur="0.2"/> so we're not going to be thinking about Bertrand <pause dur="0.8"/> and in fact if we did we know that we'd just get to the competitive price and so there'd be no price mark up and there'd be no wage mark up <pause dur="0.8"/> unless we bring in some <trunc>k</trunc> other kinds of constraints to prevent that <pause dur="0.6"/> process unravelling in that way </u><pause dur="0.2"/> <u who="sm0793" trans="pause"> mm </u><pause dur="0.3"/> <u who="nm0785" trans="pause"> yeah <pause dur="1.2"/> # <pause dur="1.4"/> okay so that i think describes <pause dur="1.3"/>

the the key assumptions <pause dur="0.5"/> and what i want us to do now is just to sketch <pause dur="0.7"/> # the method of of solving such a <trunc>s</trunc> simple model <pause dur="0.6"/> # <pause dur="1.1"/> and <pause dur="1.5"/> we've got this <kinesic desc="indicates board" iterated="n"/> two stage game going on of course we know that backward <trunc>in</trunc> by <trunc>back</trunc> we solve it by backward induction <pause dur="0.3"/> we think first about stage one <pause dur="0.8"/> reason for doing that of course is that <pause dur="0.4"/> when unions <pause dur="0.8"/> # <pause dur="0.2"/> <shift feature="voice" new="laugh"/>sorry let's think <shift feature="voice" new="normal"/>about stage two <pause dur="0.3"/> i've got i've <kinesic desc="indicates board" iterated="n"/> written these in a strange way <pause dur="0.5"/> # <pause dur="0.2"/> we're going to think first about stage two because when unions and firms <pause dur="0.3"/> are choosing their wages they need to be doing that in the context of knowing what the implication will be <pause dur="0.7"/> for the stage two decision <pause dur="0.8"/> okay <pause dur="0.2"/> so we solve first stage two <pause dur="0.3"/> which essentially means we're generating labour demand functions <pause dur="0.7"/> and it's against those labour demand functions like these <pause dur="0.7"/><kinesic desc="indicates point on board" iterated="n"/> that firms and unions are then bargaining some <pause dur="0.4"/> subgame perfect <pause dur="0.6"/> wage <pause dur="2.5"/> so that we're getting subgame perfect Nash equilibria from this <pause dur="0.8"/> okay <pause dur="0.3"/> so we first # solve stage two <pause dur="0.3"/> by backward induction <pause dur="0.4"/> that gives

us the <pause dur="0.4"/> # <pause dur="1.2"/> the labour demand curves against which <pause dur="0.3"/> the <trunc>st</trunc> the # we can analyse the stage one <pause dur="0.6"/> gain <pause dur="0.5"/> for wage determination <pause dur="0.6"/> okay so let's <pause dur="0.3"/> let's do that <pause dur="0.6"/> i'll adopt a primitive method for <pause dur="0.4"/> cleaning the board </u><gap reason="break in recording" extent="uncertain"/> <u who="nm0785" trans="pause"> so if we # then think first about stage two <pause dur="0.3"/> for employment determination <pause dur="1.3"/> i'm # <pause dur="0.3"/> i'll just go through this now in <pause dur="0.2"/> in the next <pause dur="0.5"/> four to five minutes and then we'll stop and have a break <pause dur="2.0"/> okay so stage two <pause dur="2.4"/> and you can follow this from from the handout if you if you prefer <pause dur="0.9"/> we're saying <kinesic desc="writes on board" iterated="y" dur="1"/>let's think about <pause dur="0.5"/> the decision <trunc>bay</trunc> being made by an individual firm I <pause dur="0.8"/> that firm is going to be <pause dur="0.4"/> # <trunc>tr</trunc> trying to maximize profits given by <kinesic desc="writes on board" iterated="y" dur="10"/> pi-I <pause dur="1.2"/> which of course is simply <pause dur="1.2"/> price <pause dur="0.8"/> minus <pause dur="0.5"/> wage costs <pause dur="0.2"/> of that firm I <pause dur="1.1"/> times its <pause dur="0.5"/> output level X-I which of course is also its employment level <pause dur="2.3"/> and price we know is <pause dur="0.3"/> being given by <pause dur="0.5"/><kinesic desc="indicates point on board" iterated="n"/> this linear demand function <pause dur="1.1"/> where market output is the sum of the outputs of the individual firms <pause dur="0.8"/> so we can <pause dur="0.5"/> substitute <pause dur="0.2"/><kinesic desc="indicates point on board" iterated="n"/> these <pause dur="0.3"/> two <pause dur="0.5"/> equations <pause dur="0.2"/><kinesic desc="indicates point on board" iterated="n"/>

into the profit equation <pause dur="0.7"/> and from the <pause dur="0.4"/> point of view of the handout that's giving you equation four <pause dur="1.0"/> and when we then <pause dur="0.4"/> maximize profits with respect <pause dur="0.2"/> to the choice of <kinesic desc="writes on board" iterated="y" dur="9"/> output by <pause dur="0.3"/> firm I <pause dur="0.2"/> so when we look at D-pi-I by D-X-I <pause dur="1.0"/> we get equation five <pause dur="1.1"/> which is A minus <pause dur="0.5"/> W-I <pause dur="0.4"/> A of course is the reservation price from the product demand <pause dur="0.8"/> equation so okay so always be thinking about A as the reservation price <pause dur="0.3"/> intercept on the vertical axis <pause dur="0.6"/> in a price quantity diagram <pause dur="0.8"/> # <pause dur="0.6"/> minus the wage that the firm has to pay <pause dur="2.4"/><kinesic desc="writes on board" iterated="y" dur="2"/> minus B which is the slope <pause dur="0.3"/> parameter from the market demand equation <pause dur="0.3"/> times <kinesic desc="writes on board" iterated="y" dur="12"/> the following thing <pause dur="0.3"/> X-I <pause dur="1.6"/> times D <pause dur="1.5"/> over the sum of X-Is and i'll explain this in a moment <pause dur="5.5"/> D-X-I <pause dur="0.7"/> plus the sigma <pause dur="0.5"/> X-<pause dur="0.5"/>I I going from one <pause dur="0.5"/> to N <pause dur="2.4"/> and that's the first order condition for product maximizing by the firm <pause dur="0.9"/> # <pause dur="0.8"/> in this way that's equation <pause dur="0.3"/> five from the handout <pause dur="0.9"/> okay you <pause dur="0.3"/> it's relatively straight forward to to to to demonstrate

this for yourselves i'll leave that as an exercise if anyone wants <pause dur="0.4"/> any guidance on it by all means at any point come and see me <pause dur="0.3"/> i have a generally a an open door policy for people on this course so come by my office any time you like <pause dur="0.9"/> # <pause dur="1.7"/> then what what this is saying <pause dur="0.2"/> is that <pause dur="0.6"/> # if we <pause dur="0.9"/><kinesic desc="writes on board" iterated="y" dur="1"/> if we rearrange it <pause dur="0.5"/> it's <pause dur="0.2"/> becomes easier to <pause dur="0.4"/> to see it # i guess well let's just think about it from this point of view <pause dur="2.0"/> W-<pause dur="0.5"/>I of course <pause dur="0.3"/> is the firm's <pause dur="0.3"/> marginal cost of an extra unit of output <pause dur="0.3"/> terms of having to employ an extra worker <pause dur="1.5"/> and and then the rest of the thing is giving us the <pause dur="0.2"/> the marginal <pause dur="0.4"/> # revenue <pause dur="0.4"/> to the firm from an extra unit of output <pause dur="0.8"/> and that's being made up <pause dur="0.4"/> not only of the <pause dur="0.3"/> price it's going to obtain from any <pause dur="0.5"/> increase in its own output but it has to bear in mind how a change in its output <pause dur="0.4"/> affects <pause dur="0.3"/> the output of other firms so <kinesic desc="indicates point on board" iterated="n"/> this is where the conjectural variation parameter's going to come in <pause dur="2.1"/> and let's let's see that <pause dur="0.9"/> this thing here <pause dur="0.4"/><kinesic desc="indicates point on transparency" iterated="n"/> oops i've <kinesic desc="writes on board" iterated="y" dur="1"/> missed

a bracket <pause dur="0.8"/><kinesic desc="indicates point on board" iterated="n"/><kinesic desc="writes on board" iterated="y" dur="2"/> this thing here i'm saying <pause dur="0.7"/><vocal desc="clears throat" iterated="n"/> <pause dur="0.5"/> is <pause dur="1.7"/> the # <pause dur="1.7"/> <kinesic desc="writes on board" iterated="y" dur="2"/> can be written as one plus lambda <pause dur="1.8"/> and let me just say why that is <pause dur="1.7"/><kinesic desc="writes on board" iterated="y" dur="1"/> and it's because <pause dur="0.7"/><kinesic desc="indicates point on board" iterated="n"/> this change <pause dur="0.2"/> in the sum of all the Xs can be broken down into the <kinesic desc="writes on board" iterated="y" dur="21"/> change in firm I's output <pause dur="0.9"/> plus <pause dur="0.3"/> the change <pause dur="0.4"/> in <pause dur="0.4"/> the summed output of all the other firms <pause dur="1.0"/> so let's call that the sum of the X-<pause dur="0.2"/>Js where <pause dur="0.4"/> J goes from <pause dur="0.3"/> one to N <pause dur="0.3"/> but excludes <pause dur="0.3"/> firm I itself <pause dur="1.8"/> all over D-X-I <pause dur="3.2"/> and it's clear that <pause dur="0.5"/> D-X-I over D-X-I is one <pause dur="0.2"/><kinesic desc="indicates point on board" iterated="n"/> that gives us that one here <pause dur="0.7"/> and lambda <pause dur="0.3"/> which we're defining as the conjectural variation parameter <pause dur="0.4"/> is telling us <pause dur="0.3"/> how <pause dur="0.2"/> the firm believes at least <pause dur="0.3"/> the <trunc>out</trunc> <pause dur="0.3"/> the sum of the output of all the other firms will respond <pause dur="0.6"/> to a change in firm I's output itself <pause dur="0.5"/> under the Cournot assumption <pause dur="0.2"/> lambda is zero of course because you take the output of the other firms as given <pause dur="1.2"/> okay so lambda is zero is is going to give us the

special <pause dur="0.4"/> Cournot-Nash best reply function approach <pause dur="1.7"/> okay so that's <pause dur="1.0"/><kinesic desc="indicates point on board" iterated="n"/> the first order condition for <pause dur="0.6"/> this particular firm <pause dur="1.2"/> and of course there are N <pause dur="0.2"/> such <pause dur="0.2"/> first order conditions because there are N firms so we get N first order conditions <pause dur="2.5"/> when we <pause dur="1.0"/> solve those which is actually very very straightforward <pause dur="2.7"/> okay <pause dur="0.5"/> and the easiest way <pause dur="0.5"/> to solve it is is the following <pause dur="0.2"/> when you look at <pause dur="0.5"/><kinesic desc="indicates point on board" iterated="n"/> this <trunc>in</trunc> this firm I's <pause dur="0.7"/> first order condition <pause dur="0.3"/> you'll see that <pause dur="0.4"/> what we're wanting to do is to see how its output <pause dur="0.3"/> X <pause dur="0.7"/> depends upon <pause dur="0.5"/> the vector of wages <pause dur="0.5"/> okay its output X <pause dur="0.7"/><event desc="student opens door" iterated="n" n="su0795"/> to maximize its profits <pause dur="0.7"/> we're just going to be two minutes finishing <pause dur="1.3"/> its <pause dur="0.5"/> its optimal choice of X is clearly going to depend upon its own wage as you'd expect <pause dur="0.2"/> in in a monopoly model of any model that would be true <pause dur="0.7"/> but what's happening here is that <pause dur="1.1"/> X is also going to depend <pause dur="0.2"/> X-I is going to depend <pause dur="0.7"/> # on all the X-Js <pause dur="0.7"/> and the X-Js of course will depend on the W-Js <pause dur="0.5"/><event desc="student opens door" iterated="n" n="sm0794"/> so

what we want to do is to solve this </u><pause dur="0.7"/> <u who="sm0794" trans="pause"> sorry </u><pause dur="0.2"/> <u who="nm0785" trans="pause"> just be two more minutes <pause dur="1.5"/> what we're going to be doing is <pause dur="0.2"/> to <pause dur="1.1"/> to <trunc>so</trunc> <pause dur="0.2"/> to solve this for the X-I <pause dur="0.8"/> <trunc>ex</trunc> # <pause dur="0.3"/> getting rid of all the X-Js <pause dur="0.7"/> so that X-I can be expressed in terms of <pause dur="0.5"/> of <pause dur="0.2"/> stage one information the wages okay so what we want to do is to solve this first order condition <pause dur="0.4"/> to have the profit maximizing choice of X-I <pause dur="0.4"/> expressed <pause dur="0.6"/> not in any X-Js not in any stage two outputs but in terms purely of the vector of <pause dur="0.3"/> stage one information the wage information <pause dur="0.6"/> and it's easy to do that because <pause dur="0.4"/> # <pause dur="0.7"/> the sum of all the X-Is just comes by summing <kinesic desc="indicates point on board" iterated="n"/> all of these first order conditions <pause dur="1.3"/> and then cancelling the sum of the X-Is <pause dur="0.2"/> or the at least the X-Js <pause dur="0.4"/> to leave us with X-I in terms of all the wages <pause dur="0.6"/> okay <pause dur="0.2"/> i'll leave that as an exercise for you to think about but it's actually once you've got it it's a nice little trick to do it <pause dur="0.3"/> again come and see me if you want <pause dur="0.2"/> a bit of guidance on that <pause dur="0.3"/> and that's where i'll resume <trunc>af</trunc> after a brief break