Skip to main content


<?xml version="1.0"?>

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





<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="00:45:24" n="6569">


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



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



<person id="nm0754" role="main speaker" n="n" sex="m"><p>nm0754, main speaker, non-student, male</p></person>

<person id="sm0755" role="participant" n="s" sex="m"><p>sm0755, participant, student, male</p></person>

<person id="sf0756" role="participant" n="s" sex="f"><p>sf0756, participant, student, female</p></person>

<personGrp id="ss" role="audience" size="u"><p>ss, audience, large group </p></personGrp>

<personGrp id="sl" role="all" size="u"><p>sl, all, large group</p></personGrp>

<personGrp role="speakers" size="5"><p>number of speakers: 5</p></personGrp>





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

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

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

<item n="partlevel">UG2</item>

<item n="module">unknown</item>




<u who="nm0754"> by a dominant firm i will be getting on to that <pause dur="0.5"/> in about ten minutes time <pause dur="0.7"/> so if you could have that # <pause dur="0.3"/> handy for then <pause dur="0.9"/> # <pause dur="1.7"/> and i'll assume everybody has now returned <pause dur="4.4"/> i hope you <pause dur="0.7"/> might recall that the last time i gave a lecture on this which was <pause dur="0.2"/><kinesic desc="puts on transparency" iterated="n"/> this was eleven o'clock last <pause dur="0.4"/> Friday <pause dur="0.3"/> i was talking about market structure <pause dur="0.6"/> in particular entry conditions and i had got really down to <pause dur="0.9"/> the last part of this subheading here a <trunc>ne</trunc> the new industrial organization approach <pause dur="0.6"/> what i want to do this morning is first of all to <pause dur="1.2"/> finish that off and also to say something about <pause dur="0.5"/> perfectly contestable markets <pause dur="1.1"/> all under the general <pause dur="0.5"/> heading of entry conditions <pause dur="1.2"/> and then to go on <pause dur="0.8"/> as i say in about ten minutes time to a very large topic <pause dur="0.7"/> # topic five in the course outline pricing decisions but i'll introduce that <pause dur="0.7"/> # when i get there <pause dur="1.8"/> you may recall that <pause dur="0.7"/> what i said at the end of the <trunc>la</trunc> of my last lecture <pause dur="0.2"/> on the new industrial organization approach particular to particularly to <pause dur="0.5"/> # towards entry conditions and market performance <pause dur="0.5"/> was that they had laid <pause dur="0.9"/> they're essentially gain theories they had laid <pause dur="0.9"/> # <pause dur="0.3"/> considerable <pause dur="0.2"/> <trunc>o</trunc> <pause dur="0.9"/> emphasis on <pause dur="0.4"/> commitment of resources by the incumbent firms <pause dur="1.2"/> in relation to new entrants or possible entrants <pause dur="0.5"/> and the credible threats <pause dur="0.4"/> to <pause dur="0.4"/> the possible position of any <pause dur="0.3"/> entry that took place that these commitments actually <pause dur="0.4"/> made <pause dur="0.8"/> if a firm simply <trunc>threat</trunc> an incumbent firm simply threatened <pause dur="0.5"/> # a potential entrant that it would do a number of things to prevent or frustrate its entry <pause dur="0.5"/> and there was not that commitment there <pause dur="0.2"/> then the threat would be empty and it would probably be ignored <pause dur="0.4"/> by <pause dur="0.3"/> the potential entrant and they the the entry will actually occur <pause dur="0.3"/> to the disadvantage of the incumbent firm <pause dur="0.8"/> i will give some examples

<pause dur="0.8"/> more specific examples of that # <pause dur="0.3"/> as i get into topic # five on pricing but for the moment if you simply <pause dur="0.4"/> note <pause dur="0.4"/> that the new industrial organization approach <pause dur="0.4"/> put particular emphasis on this notion of commitment <pause dur="0.3"/> and that that commitment then offered <pause dur="0.3"/> credible threats <pause dur="0.2"/> to potential entrants <pause dur="0.7"/> my final point about this approach under <pause dur="0.5"/> three was that they were or <trunc>hi</trunc> they continue to be highly critical of the Chicago school <pause dur="0.2"/> approach to the <trunc>a</trunc> analysis of market structure <pause dur="0.2"/> and performance <pause dur="0.7"/> in particular <pause dur="0.8"/> this group <pause dur="0.2"/> thinks that the <trunc>chicag</trunc> the implicit model used by the Chicago school <pause dur="1.3"/> is oversimplified <pause dur="0.9"/> implicitly in much of their <trunc>ma</trunc> of the Chicago school's market analysis is the theory of perfect competition <pause dur="0.8"/> and the new industrial organization school regards this as highly <pause dur="0.3"/> suspect for the purposes for which it's put <pause dur="0.5"/> # by the Chicago school <pause dur="1.5"/> an extension to that is that they <pause dur="0.7"/> argue that is the new industrial organization <pause dur="0.6"/> # group <pause dur="0.4"/> argue that the Chicago school <pause dur="1.3"/> ignores the <pause dur="0.4"/> # <pause dur="0.2"/> fact that in most real world markets information <pause dur="0.4"/> is highly imperfect <pause dur="1.3"/> and that that imperfect information <pause dur="0.5"/> either particularly on the part of potential entrants that they would have far less information available to them <pause dur="0.4"/> than is available to <pause dur="0.3"/> the incumbent firms <pause dur="1.5"/> that imperfect information <pause dur="0.2"/> may make strategies on the part of incumbent firms <pause dur="0.8"/> much more plausible <pause dur="0.4"/> than if you are operating or theorizing in a world of perfect information <pause dur="0.2"/> if everybody's got perfect information <pause dur="0.4"/> the Chicago school <pause dur="0.3"/> argument may make more sense <pause dur="0.8"/> in real world markets that information <pause dur="0.5"/> # is far from perfect <pause dur="0.6"/> and <pause dur="0.4"/> # that may therefore make <pause dur="0.2"/> it possible and logical for incumbent firms to pursue a number of strategies <pause dur="0.5"/> # to deter <pause dur="0.3"/> or coerce <pause dur="0.5"/> potential and actual <pause dur="0.3"/> entrants <pause dur="0.2"/> and again i will <pause dur="0.5"/> refer back to that point when i'm talking about various aspects of pricing policies <pause dur="0.3"/> which i'm about to come on to <pause dur="1.2"/> so they are highly critical of the Chicago school particularly for its <pause dur="0.5"/>

oversimplified theoretical approach <pause dur="0.4"/> for their <pause dur="0.2"/> # assumption that there is usually <pause dur="0.5"/> something like perfect information <pause dur="0.2"/> and if you drop those <pause dur="0.2"/> assumptions then you can come up with quite <pause dur="0.3"/> different conclusions <pause dur="1.0"/> finally on on this # <pause dur="0.7"/> # topic of entry conditions i want to mention <pause dur="0.5"/> a theory that i'm sure a number of you are already familiar with from other courses <pause dur="0.6"/> the theory of perfectly contestable markets note the word contestable <pause dur="0.3"/> not <pause dur="0.4"/> competitive <pause dur="1.1"/> in fact one of the points that i will <pause dur="0.5"/> draw to your attention if you're not already aware of it is that <pause dur="0.4"/> this theory the new one <pause dur="1.3"/> <vocal desc="cough" iterated="n"/> comes to the same conclusions as far as the performance of the markets are concerned <pause dur="0.3"/> as the perfectly competitive model <pause dur="1.0"/> but it has some quite <pause dur="0.5"/> # different <pause dur="0.2"/> # <pause dur="1.1"/> prediction <trunc>w</trunc> <trunc>i</trunc> it the the the structure of of the markets involved are quite different <pause dur="0.3"/> from those involved <pause dur="0.4"/> in the perfectly competitive model <pause dur="2.9"/> perfectly <trunc>conte</trunc> a perfectly contestable market <pause dur="0.3"/> is one where there are zero entry barriers and zero exit costs <pause dur="0.7"/> it's that emphasis <pause dur="0.5"/> on zero entry barriers is which is why i particularly thought it appropriate to mention it <pause dur="0.3"/> in the present context <pause dur="1.9"/> the notion of zero exit barriers <pause dur="1.2"/> implies that the any entrant firm to that market incurs zero sunk costs <pause dur="2.0"/> so a firm <pause dur="0.7"/> <trunc>th</trunc> <trunc>th</trunc> <trunc>th</trunc> the theory isn't assuming that a firm doesn't actually have to incur production costs if it enters the market but what it is <vocal desc="cough" iterated="n"/> <pause dur="0.4"/> in its perfect form assuming <pause dur="0.6"/> is that an entrant can come in <pause dur="0.6"/> and produce

without incurring <pause dur="0.3"/> any sunk costs there are zero <pause dur="0.2"/> entry barriers <pause dur="0.3"/> that <pause dur="0.6"/> theory in its perfect form has quite dramatic <pause dur="0.5"/> predictions <pause dur="1.4"/> in particular it predicts that the properties of a perfectly contestable market will be exactly the same as those of a perfectly <pause dur="0.2"/> competitive market <pause dur="2.6"/> however it predicts that those results will hold even if you've only have <pause dur="0.5"/> two or three firms operating in the market <pause dur="1.2"/> now that of course is quite dramatically different from what you predict from the theory of perfect competition <pause dur="1.8"/> the theory of perfectly contestable markets is saying that if you have free entry and free exit <pause dur="1.0"/> you will have in equilibrium <pause dur="0.5"/> prices equal to marginal costs you will have all firms producing at minimum unit costs <pause dur="0.7"/> they will <pause dur="0.3"/> the industry output will be produced at minimum cost and they will <pause dur="0.4"/> the market would clear <pause dur="1.1"/> all consumers will be satisfied at a price which is equal to marginal cost <pause dur="2.2"/> even if you only have two firms <pause dur="0.3"/> or three or four firms in the market <pause dur="0.8"/> now that <pause dur="0.4"/> seems at first glance perhaps counter-intuitive <pause dur="0.6"/> because one has <pause dur="0.2"/> normally <trunc>gr</trunc> <trunc>gr</trunc> grown up with the idea that if you have a highly concentrated market that is consisting of only two or three firms <pause dur="0.4"/> they will in fact <pause dur="0.2"/> be able to wield considerable market power <pause dur="1.2"/> what is driving the whole of that result <pause dur="0.9"/> in <trunc>perfe</trunc> the the perfectly <trunc>con</trunc> <pause dur="0.6"/> # testable <pause dur="0.3"/> market <pause dur="0.3"/>

# theory what it drives the entire result <pause dur="0.4"/> is the free entry and the free exit <pause dur="0.6"/> without going into too much detail <pause dur="0.9"/> essentially what # the theory <pause dur="0.2"/> predicts is that if the an incumbent firm in such a market tries to raise its price above marginal cost <pause dur="0.6"/> an entrant can <trunc>m</trunc> immediately appear <pause dur="0.5"/> undercut <pause dur="0.3"/> that price <pause dur="0.5"/> so long as it's in excess of marginal cost <pause dur="0.3"/> make a profit <pause dur="0.4"/> if the incumbent firm responds <pause dur="0.4"/> to that price cut <pause dur="1.8"/> if the incumbent for example drops its own price to marginal cost <pause dur="0.7"/> then the <pause dur="0.4"/> new firm having made a profit <pause dur="0.3"/> previously <pause dur="0.3"/> can then leave costlessly <pause dur="2.1"/> the knowledge on the part of the incumbent firm <pause dur="0.4"/> that that is the case that is if it tries to rise its price or if there are two or three or firms <pause dur="0.4"/> incumbent firms if they try <trunc>t</trunc> try to raise their price <pause dur="0.7"/> it would immediately provoke entry <pause dur="1.0"/> and the price will then sink to marginal cost <pause dur="0.3"/> will mean that the incumbent firms will be unable to raise their price <pause dur="1.0"/> above <pause dur="0.4"/> # <pause dur="1.6"/> marginal cost <pause dur="0.8"/> the significance therefore of the notion of perfectly free <pause dur="0.6"/> entry and exit <pause dur="0.6"/> you can see it's i hope it's significant that here was a theory which was saying <pause dur="1.3"/> even if you've got very highly concentrated

oligopolies <pause dur="1.1"/> if these conditions hold <pause dur="0.4"/> then you needn't worry there are there are very few policy <pause dur="0.4"/> policies you need adopt towards such industries because they will <pause dur="0.3"/> produce a performance which is <pause dur="0.3"/> in line with that of a perfectly competitive market <pause dur="0.8"/> the problem with the theory <pause dur="1.9"/> <vocal desc="cough" iterated="n"/> is well sorry before i get on to that let me simply emphasize <pause dur="0.5"/> that the theory first made its appearance in the late seventies by the early eighties <pause dur="0.5"/> it had <pause dur="0.4"/> it had already been fed in to policy decisions in the United States and by the mid-eighties <pause dur="0.5"/> it was having an influence on policy decisions in Britain as well <pause dur="0.6"/> in a white paper in the middle of nineteen-eighties for example <pause dur="0.4"/> on the <pause dur="0.5"/> # deregulation of the local bus markets <pause dur="0.2"/> the notion of a contestable market <pause dur="0.5"/> was <pause dur="0.4"/> # actually used and it was discussed at some length the argument being that these markets were <pause dur="0.3"/> contestable and therefore <pause dur="0.3"/> # <pause dur="0.4"/> they would be put in a competitive performance <pause dur="0.5"/> notice <pause dur="0.7"/> i there said contestable <pause dur="0.2"/> rather than perfectly contestable <pause dur="0.7"/> the problem <pause dur="0.5"/> as it arose in the debate in the <trunc>uni</trunc> over this in the United States <pause dur="0.6"/> was that a number of policy makers were saying well <pause dur="0.4"/> these markets are approximately contestable <pause dur="0.6"/>

therefore there is no need for us to worry <pause dur="0.3"/> particularly in civil aviation which the Americans were deregulating at that time <pause dur="0.9"/> a lot of mergers were proposed <pause dur="1.1"/> under the influence essentially of this theory of perfectly <trunc>con</trunc> perfect contestability <pause dur="0.5"/> the authorities then said well <pause dur="0.2"/> although this <pause dur="0.3"/> these mergers will concentrate the industry or <trunc>the</trunc> they will concentrate these particular <pause dur="1.2"/> parts of the <pause dur="0.4"/> of the market <pause dur="1.4"/> we will not intervene to try to prevent these mergers or look at them more closely because we are advised that these are contestable markets if you <pause dur="0.5"/> if an entrant <pause dur="0.3"/> <trunc>i</trunc> if they try to raise their price <pause dur="0.4"/> after the merger <pause dur="0.2"/> entrants would come in and drive the price down <pause dur="2.0"/>

# <pause dur="1.2"/> so <pause dur="0.5"/> it's a relatively recent theory it has an almost <trunc>im</trunc> <pause dur="0.5"/> # immediate impact on policy discussions both in the United States and in Britain and elsewhere <pause dur="0.5"/> the notion of contestability has passed very much into the economic literature <pause dur="0.9"/> the problem with it all however is that the theory is non-robust <pause dur="0.9"/> i'm using that term <pause dur="0.4"/> in its technical sense <pause dur="1.2"/> any scientific theory to be robust <pause dur="1.2"/> has to have the following characteristic <pause dur="1.2"/> that if you <pause dur="0.8"/> if the assumptions of the theory <pause dur="1.3"/> do not hold completely <pause dur="0.2"/> or perfectly <pause dur="0.2"/> but they hold approximately <pause dur="1.7"/> then for a theory to be robust <pause dur="0.2"/> under those circumstances the predictions <pause dur="0.5"/> must only change slightly <pause dur="1.0"/> you could say that the theory of perfect competition is a robust theory because <pause dur="0.3"/> if the assumptions <pause dur="0.3"/> are not met completely but approximately in such a theory <pause dur="0.4"/> the results don't change very much <pause dur="1.0"/> you still get <pause dur="0.6"/> prices approximately in line with marginal cost <pause dur="0.4"/> in the theory of perfect competition <pause dur="0.7"/> if the assumptions don't hold completely <pause dur="0.9"/> so a robust theory is one where if the <pause dur="0.2"/> assumptions don't hold perfectly but approximately the predictions don't change very much <pause dur="1.3"/> my <pause dur="0.4"/> # <pause dur="0.2"/> critique or # <trunc>th</trunc> the critique of others as well <pause dur="0.3"/>

about the theory of perfect contestability <pause dur="0.6"/> is that if you change the assumptions slightly <pause dur="1.5"/> the predictions change dramatically it's very unstable <pause dur="2.0"/> let me give you an example of how <pause dur="0.6"/> what i mean by that <pause dur="2.1"/> if <trunc>fo</trunc> let us say in a particular market which a number of people have said is contestable <pause dur="0.8"/> if <pause dur="0.7"/> there are inevitable delays <pause dur="1.1"/> between a firm announcing that it's coming into the market and it actually <pause dur="0.5"/> managing to <pause dur="0.2"/> produce <pause dur="2.4"/> and if in coming into the market <pause dur="0.6"/> the <trunc>f</trunc> the entrant has to incur <pause dur="0.4"/> some sunk costs they can only be they only need be slight <pause dur="0.2"/> sunk costs <pause dur="1.1"/> so if there's a delay a slight delay between saying i will come into the market the firm has to build up its capacity <pause dur="1.1"/> there's a delay between the announcement and the actual production <pause dur="0.5"/> and also <pause dur="0.4"/> if <pause dur="1.4"/> there are some slight exit costs sunk costs <pause dur="0.6"/> incur for the firm has to incur to come into the market <pause dur="0.6"/> then the predictions of the the model <pause dur="0.7"/> are dramatically different <pause dur="1.7"/>

an incumbent firm in such a market <pause dur="0.4"/> can charge the monopoly price <pause dur="0.5"/> or if it's two or three firms they can charge <pause dur="0.3"/> near the monopoly price <pause dur="0.8"/> they can charge near the monopoly price <pause dur="0.7"/> until the entrant appears <pause dur="1.0"/> they can then immediately drop their price to marginal cost <pause dur="1.6"/> the entrant <pause dur="0.5"/> having <pause dur="0.2"/> finally come in to production <pause dur="0.9"/> would then make no money <pause dur="0.3"/> in fact it would make a loss it would make a loss equal to its sunk costs <pause dur="1.6"/> if the entrant is aware of that <pause dur="1.3"/> it would not come into the market <pause dur="2.0"/> so the sequence is this that <pause dur="0.8"/> with slight alterations <trunc>i</trunc> if the <pause dur="0.6"/> the notion of a perfectly contestable market is not met if you make slight changes in the assumptions even though the market may be approximately <pause dur="0.9"/> contestable <pause dur="0.6"/> it may has a dramatic difference in the prediction because it means that the incumbent firms <pause dur="0.2"/> will continually be able to charge <pause dur="0.3"/> something approaching a monopoly price <pause dur="0.3"/> entry will not occur because the entrants will say <pause dur="0.8"/> i will have to incur slight sunk costs to get into this market <pause dur="0.5"/> and i won't be able to recover them and i won't make any money because as soon i appear and produce <pause dur="0.3"/> the price will collapse to marginal cost <pause dur="2.0"/> now <pause dur="1.2"/> the emphasis in this theory on sunk costs i think is important that's possibly its most important contribution <pause dur="1.4"/> i'm really i suppose offering you a warning that whenever you see <pause dur="0.5"/> # contestability or contestable markets mentioned in the literature <pause dur="0.3"/> ask yourself the question <pause dur="0.4"/> is this a fudge <pause dur="1.9"/> is the person talking about <pause dur="0.2"/> market X as being contestable do do they mean that it's approximately <pause dur="0.4"/> contestable in which case <pause dur="0.5"/> you need to ask the further question what does

that actually mean does it mean that the firms are able to charge a near monopoly price <pause dur="1.3"/> because <pause dur="0.2"/> as i've indicated <pause dur="0.4"/> the theory is non-robust and <pause dur="0.4"/> # <pause dur="0.3"/> the predictions <pause dur="0.6"/> are <pause dur="0.4"/> or can be very dramatically different if you alter the <pause dur="0.2"/> assumptions only slightly <pause dur="0.3"/> from the perfect <pause dur="0.2"/> case <pause dur="0.7"/> the reason for emphasizing that <pause dur="0.6"/> # <pause dur="0.4"/> is that it has had a very <trunc>im</trunc> wide impact on the literature i'm sure that you've <pause dur="0.4"/> in your microeconomics for example you may well have encountered discussions of this <pause dur="1.0"/> as a <pause dur="0.4"/> an abstract theory the the notion of perfect contestability works <pause dur="0.4"/> but as soon as you move away from the <pause dur="0.5"/> perfect the the the <pause dur="0.2"/> <vocal desc="cough" iterated="n"/> perfect version of it <pause dur="0.8"/> the predictions <pause dur="0.2"/> go all over the place <pause dur="1.6"/> entry conditions therefore are quite crucial <pause dur="0.3"/> in the predictions <pause dur="0.3"/> <kinesic desc="changes transparency" iterated="y" dur="3"/> for <pause dur="0.3"/> market <pause dur="0.8"/> conduct and market performance <pause dur="0.9"/>

what i want to <kinesic desc="changes transparency" iterated="y" dur="1"/> do now is to go on to a very large topic indeed <pause dur="0.3"/> this is topic five <pause dur="0.7"/> on <pause dur="0.6"/> page two of the course outline <pause dur="0.9"/> pricing decisions <pause dur="2.5"/> and <pause dur="0.2"/> it is a very large topic and i want to <vocal desc="cough" iterated="n"/> <pause dur="2.0"/> deal with it under two main headings i'm assuming that you have <pause dur="1.1"/> a pretty good knowledge of let us say the theory of simple monopoly the theory of perfect competition <pause dur="1.5"/> i'm going to be dealing with <pause dur="0.4"/> essentially with two forms of imperfect competition <pause dur="2.4"/> and i'm going to <pause dur="3.2"/><kinesic desc="reveals covered part of transparency" iterated="n"/> divide that up essentially into <pause dur="3.7"/> a market structure which i will characterize as being <pause dur="1.9"/> a structure with one dominant firm <pause dur="0.5"/> and i'll say a word in a moment about what i mean with <pause dur="0.2"/> by dominance as opposed to monopoly <pause dur="1.4"/> there are <trunc>var</trunc> a number of variations on the theme of pricing in dominant markets which i've listed here <pause dur="0.3"/> all of this incidentally will take me quite a long while to get through so there's no need to make a note of <pause dur="0.3"/> all of it now <pause dur="0.4"/> i think perhaps if you went down to there <kinesic desc="indicates point on transparency" iterated="n"/> for the moment <pause dur="0.4"/> that would cover what i'm going to say this morning <pause dur="1.4"/> # <pause dur="7.1"/> so the first market structure <pause dur="2.1"/> concerns <pause dur="1.8"/> not monopoly but something approaching that <pause dur="0.6"/> # a market structure with a dominant firm <pause dur="0.3"/> the second one <trunc>w</trunc> the second group of <pause dur="0.6"/> # <pause dur="0.3"/> markets that i want to talk about which is not on this # slide but which will be on another one <pause dur="0.7"/> is oligopoly <pause dur="0.4"/> which you're <trunc>a</trunc> familiar with i'm sure from microeconomics in a general way <pause dur="0.4"/> i will make <pause dur="0.8"/> # that much more precise when i come on to it but it forms the second group <pause dur="0.3"/> of <pause dur="0.3"/> topics within this general <pause dur="0.4"/> under this general heading <pause dur="0.4"/> of pricing decisions which if you look at the course outline <pause dur="0.4"/> gives under the subheading two pricing problems in oligopoly <pause dur="2.4"/> what do i then mean by the <trunc>di</trunc> or making the distinction between dominance and monopoly we all know what a monopoly is that it's a

single firm case where <pause dur="1.2"/> the firm is in sole control of a market and it's protected by <pause dur="0.4"/> such high entry barriers that its position is <pause dur="0.6"/> # <pause dur="1.0"/> not vulnerable to competition <pause dur="0.2"/> and the <trunc>f</trunc> and we can then analyse <trunc>i</trunc> from the simple theory of monopoly what the <pause dur="0.3"/> predictions about price and output will be <pause dur="1.2"/> as i've said in some contexts # <pause dur="2.7"/> already <pause dur="4.3"/> # <pause dur="2.6"/> it's very rare in the real world for firms to be in that happy position of being a complete monopoly <pause dur="0.6"/> in the real world you very often however <pause dur="0.2"/> have an approximation <pause dur="0.3"/><event desc="looks through transparencies" iterated="y" dur="14"/> to <pause dur="1.1"/> dominance <pause dur="0.9"/> # <pause dur="0.6"/> for example <pause dur="6.3"/> seem to have lost my example at the moment oh here we are <pause dur="2.7"/> i undertook a study in the middle eighties from # a large number of monopolies as it was then called the monopolies and mergers commission <pause dur="0.4"/> and i was quite it was quite easy for me to find twenty-two markets this # the the period covered was the middle seventies to the <pause dur="0.2"/> middle eighties <pause dur="0.8"/> it was quite easy to find a number of markets that they had investigated where the first firm had <pause dur="0.8"/> a share of fifty per cent or above in some cases it was much higher than that it was sort of eighty or ninety per cent <pause dur="0.9"/> and the second largest firm or firms were were more than <trunc>ha</trunc> <pause dur="0.6"/> were only half or less of the size in terms of market share <pause dur="0.3"/> of the dominant firm <pause dur="1.4"/> so although in many cases <pause dur="0.3"/> # in the real world you don't have monopoly you only have those usually in the case of natural

monopoly <pause dur="0.5"/> the notion of dominance as i want to use it <pause dur="0.2"/> is quite frequent <pause dur="1.0"/> you do <pause dur="0.3"/> frequently find <pause dur="0.2"/> one firm with a very sizeable market share <pause dur="0.3"/> as a rule of thumb if you like <pause dur="0.6"/> upwards of fifty per cent of the market possibly much higher than that <pause dur="0.6"/> and where the second largest firm is much smaller <pause dur="0.4"/> with a a share of <pause dur="0.3"/> perhaps under ten per cent or a number of firms <pause dur="0.3"/> all of whom have quite <pause dur="0.2"/> small <pause dur="0.5"/> # market shares <pause dur="0.8"/> now my <pause dur="0.2"/> # i will therefore mean by dominance that sort of market structure <pause dur="0.4"/> the size distribution of firms is highly skewed you've got one firm <pause dur="1.3"/> in pretty much in command of the market but a number of other firms <pause dur="1.3"/> operating in the market <pause dur="0.3"/> in competition <pause dur="1.5"/> that <pause dur="0.4"/> market structure can be <pause dur="0.3"/> modelled by <pause dur="0.5"/> the dominant firm and the competitive fringe <pause dur="0.2"/> case <pause dur="1.6"/> an example of which i have given round as a diagram and which i want to say something <pause dur="0.5"/> # about now <pause dur="7.8"/> <kinesic desc="changes transparency" iterated="y" dur="4"/> the essential questions that i want to address in talking about this model are the following <pause dur="0.2"/> and this is exactly the <pause dur="0.3"/> diagram that you should have in front of you <pause dur="2.2"/> the sort of questions i want to address are these <pause dur="1.9"/> to what extent is a firm <pause dur="0.3"/> in this dominant position to what extent is it constrained in its pricing behaviour <pause dur="2.2"/> by the presence of the the smaller fringe of firms <pause dur="1.0"/> and what strategies does it have for trying to control

that fringe to its own advantage <pause dur="3.1"/> right well this <pause dur="0.4"/> diagram which may at first glance look extremely complicated does <pause dur="0.3"/> or can be broken down into <trunc>s</trunc> i hope <pause dur="0.2"/> some fairly straightforward components <pause dur="1.1"/> we start out with the line M-D which stands for market demand <pause dur="1.9"/> the related marginal revenue curve or line is denoted <pause dur="0.3"/> M-R <pause dur="3.3"/> the <trunc>l</trunc> the horizontal line <pause dur="0.3"/> M-C-D is the marginal costs of the dominant firm <pause dur="2.6"/> so you've got a market demand curve its related marginal revenue curve <pause dur="0.2"/> treating <pause dur="0.6"/> or <pause dur="0.2"/> on the assumption for the moment that we have a monopoly <pause dur="2.6"/> our dominant firm's marginal costs are horizontal <trunc>a</trunc> at # M-C <pause dur="0.4"/> but they are made horizontal for to make the diagram rather simpler than it would be if we had a curve <pause dur="0.2"/> doesn't actually affect <pause dur="0.3"/> the result <pause dur="1.0"/> now let's start with the position let's suppose that this is not a dominant firm it's a monopoly <pause dur="1.4"/> if it was to charge a simple monopoly price <pause dur="0.3"/> then on well known principles it would seek out this equality here between marginal revenue and marginal cost <pause dur="0.4"/> and the monopoly price is P-M <pause dur="1.6"/> and as a monopolist it would charge <pause dur="0.6"/> Q-M for that output <pause dur="2.9"/>

now that's if you like the starting point <pause dur="0.2"/> remember <trunc>on</trunc> one of the questions i wanted to pose of this model is well <pause dur="0.2"/> how is the behaviour of such a firm constrained by the presence of a fringe <pause dur="0.2"/> small much smaller competitors <pause dur="2.5"/> we've now got a line <pause dur="0.3"/> # <pause dur="0.3"/> an upward sloping line marked S-F which <trunc>s</trunc> stands for the supply by small fringe firms <pause dur="3.1"/>

that is upward sloping for <pause dur="0.3"/> normal supply reasons that is if the price is higher it <trunc>th</trunc> <trunc>th</trunc> <pause dur="0.3"/> <trunc>th</trunc> the price that the output can command is higher more will be <trunc>supri</trunc> supplied <pause dur="2.0"/> we can note as well that if we look at the point P-zero or P-O <pause dur="0.3"/> on the vertical axis <pause dur="1.5"/> at a price below that the fringe doesn't supply anything at all <pause dur="1.7"/> in other words the fringe <pause dur="0.6"/> fringe's costs are higher <pause dur="0.9"/> than the costs of the dominant firm by assumption <pause dur="3.7"/> at a price P-zero or below they will supply nothing <pause dur="1.4"/> let's look at a <trunc>pr</trunc> at # <pause dur="0.3"/> the <pause dur="1.0"/> the equivalent point <pause dur="0.5"/> on the demand curve opposite P-F

in this diagram <pause dur="0.3"/> which i'm <pause dur="0.6"/> showing by the position of my pencil at the moment <pause dur="1.0"/> we trace that through onto the demand curve if the price rose to P-F <pause dur="1.0"/> the implications from basic analysis is that the fringe at that price would be supplying the whole <pause dur="0.3"/> output the whole market it would be supplying enough to clear the market <pause dur="0.3"/> at the price <pause dur="0.3"/> P-F <pause dur="2.7"/> for our purposes therefore the relevant price range <pause dur="0.4"/> is P-F<pause dur="0.6"/>P-0 <pause dur="0.5"/> over that range <pause dur="0.9"/> the fringe <pause dur="1.5"/> supplies part <pause dur="0.5"/> or all in the limit <pause dur="0.5"/> of the market <pause dur="1.6"/> and it's upward sloping as i've indicated because the <trunc>pri</trunc> the higher the price the more will be supplied <pause dur="1.6"/> we make the further assumption now given this structure that we if we had our dominant firm <pause dur="0.7"/> with the fringe <pause dur="0.8"/> the dominant firm <pause dur="1.7"/> makes the price <pause dur="1.3"/> it optimizes given the presence of that fringe <pause dur="0.8"/> the fringe suppliers because they're so small by assumption <pause dur="0.3"/> simply take that as a <trunc>para</trunc> parametric price they can't <trunc>affor</trunc> # can't # affect it <pause dur="0.9"/> they will simply supply according to that <pause dur="0.6"/> # as if you if you

like as a fixed market price <pause dur="0.4"/> depending on what price has been fixed <pause dur="0.4"/> by the dominant firm <pause dur="2.1"/> now <pause dur="0.7"/> what how therefore does under those assumptions and this # market structure how does the dominant firm determine its price and output <pause dur="0.8"/> well the line that i have marked in this diagram as R-D which is this line here <pause dur="2.2"/><kinesic desc="indicates point on transparency" iterated="n"/> this is that stands for residual demand <pause dur="1.6"/> and it is simply the market demand <pause dur="0.4"/> minus <pause dur="0.9"/> whatever amount at the price specified is supplied by the fringe suppliers <pause dur="2.3"/> the line R-D <pause dur="1.5"/> is residual demand <pause dur="1.3"/> we are taking away from the market demand <pause dur="0.5"/> the amount supplied by the fringe <pause dur="0.3"/> suppliers <pause dur="0.3"/> to leave us with a residual demand R-D <pause dur="2.6"/> that is the demand <pause dur="0.5"/> given the presence of the fringe <pause dur="0.5"/> that is the demand on which the dominant firm optimizes <pause dur="3.4"/> related to that curve or line R-D is the residual marginal revenue curve M-R-R in the diagram <pause dur="3.6"/> so we've got the dominant firm <pause dur="0.7"/> taking account of the fringe suppliers <pause dur="0.8"/> it goes through or in in <pause dur="0.5"/> # in principle <pause dur="0.4"/> it goes through the mental process of

actual subtracting from the market demand <pause dur="0.2"/> what <pause dur="0.4"/> the fringe will supply and then optimizes <pause dur="0.5"/> # <pause dur="0.8"/> on that residual demand curve <pause dur="0.4"/> we can then proceed <pause dur="0.4"/> to see what the the dominant firm does this is still its marginal cost <pause dur="0.4"/> this is now its marginal revenue <pause dur="0.6"/> so the relevant intersection <pause dur="0.4"/> is this point here <pause dur="3.3"/><kinesic desc="indicates point on transparency" iterated="n"/> on the residual demand curve that implies because it's directly above the intersection of marginal cost and marginal revenue <pause dur="0.4"/> that implies <pause dur="0.2"/> a price of P-R <pause dur="5.4"/> the dominant firm is optimizing given its marginal cost <pause dur="0.7"/> and given its relevant marginal revenue which is M-R-R <pause dur="0.3"/> the relevant intersection is at this point here <pause dur="3.0"/> <kinesic desc="indicates point on transparency" iterated="n"/> reading off from the demand curve <pause dur="0.2"/> the optimizing for the monopolies the optimizing price <pause dur="0.5"/> is P-R <pause dur="2.1"/> you may well say well the market won't the whole market won't clear at that price <pause dur="0.5"/> because if this firm is only supplying Q-R at that which it would do the dominant firm <pause dur="0.4"/> is only supplying Q-R <pause dur="1.5"/> at that price an amount Q-S is demanded <pause dur="1.1"/> well of course the gap <pause dur="0.9"/> this <pause dur="0.4"/> dashed line

here <pause dur="1.3"/><kinesic desc="indicates point on transparency" iterated="n"/> the gap is filled <pause dur="0.3"/> by <pause dur="1.0"/> the fringe <pause dur="0.3"/> they <pause dur="0.3"/> account for <pause dur="0.3"/> an amount of the supply Q-R-Q-S <pause dur="2.0"/> and the market will clear at the price P-R <pause dur="0.6"/> so what we've got <pause dur="0.5"/> is the dominant firm producing Q-R <pause dur="1.0"/> optimizing its position <trunc>i</trunc> its position <pause dur="0.3"/> given the constraint <pause dur="0.7"/> offered by the fringe suppliers <pause dur="1.4"/> the fringe suppliers will <pause dur="0.7"/> take that market that that price has given <pause dur="0.3"/> they're not going to try to effect it because they're too small in relation to the total <pause dur="0.4"/> they supply Q-R-Q-S <pause dur="3.8"/> what we can note about this is compared with the <pause dur="2.1"/> simple monopoly case <pause dur="0.7"/> where we had a price of P-M and an output Q-M <pause dur="0.5"/> the price <pause dur="1.1"/> because of the presence of <pause dur="0.5"/> apparently insignificant small firms the price is <pause dur="0.3"/> quite different the price has dropped to P-R <pause dur="2.2"/> and the amount supplied by the fringe is is as i've indicated so the price is quite <trunc>amoun</trunc> an amount lower <pause dur="0.4"/> as a result of the presence of the fringe <pause dur="0.3"/> and of course the amount sold <pause dur="0.4"/> is lower <pause dur="0.5"/> consumers therefore undoubtedly benefit from the presence of the fringe even

though <pause dur="0.7"/> they may appear on the face of it to be <pause dur="0.2"/> rather <pause dur="0.2"/> insignificant <pause dur="2.1"/> let me stop there for a moment and ask if anybody <pause dur="0.4"/> wants me to go through any of that again because i know at # first <pause dur="0.4"/> this is your first <pause dur="1.1"/> look at that it it's quite complicated <pause dur="0.6"/> does anybody want me to go through any of it again <pause dur="1.9"/> yes can you pinpoint <pause dur="0.2"/> where you would like me to <pause dur="0.3"/> to go through </u> <pause dur="0.4"/> <u who="sm0755" trans="pause"> yeah just <pause dur="0.2"/> one small bit i didn't catch <pause dur="0.2"/> why the range between P-nought and <pause dur="0.5"/> P-F i understand the lower range being P-nought but not <pause dur="0.2"/> why the upper range should be P-F </u><pause dur="1.3"/> <u who="nm0754" trans="pause"> well <trunc>o</trunc> the the upper range <pause dur="0.3"/> # it where the supply curve of the fringe cuts the demand curve <pause dur="0.4"/> on basic <pause dur="0.3"/> supply and demand principles <pause dur="0.3"/> the fringe would simply be <pause dur="0.3"/> would would sell <pause dur="0.2"/> # <pause dur="1.1"/> well it <trunc>a</trunc> it it looks as if it's <trunc>a</trunc> almost identical with Q-R but that was an unintentional <pause dur="0.5"/> the fringe would supply the entire market demand at a price as high as P-F <pause dur="1.7"/> and the lower level as is as you've <pause dur="0.4"/> indicated is simply <pause dur="0.2"/> that as its costs are higher <pause dur="0.9"/> at a price P <trunc>no</trunc> P-<pause dur="0.4"/>O or below it doesn't supply anything <trunc>a</trunc> they they don't supply anything at all <pause dur="1.4"/> yes </u><u who="sf0756" trans="latching"> just ask what was residual demand is that for <gap reason="inaudible" extent="2 secs"/></u><u who="nm0754" trans="latching"> no <pause dur="0.2"/> it's <pause dur="0.2"/> it's <pause dur="0.2"/> the the <trunc>do</trunc> the # the mental process we assume that the

the dominant firm goes through is this <pause dur="0.4"/> i know roughly what the market demand is <pause dur="0.7"/> i know what the fringe will supply <pause dur="1.4"/> i will subtract from the market demand what the fringe will supply if i charge <pause dur="0.8"/> this price <pause dur="0.7"/> and then i will optimize <pause dur="0.2"/> on that <pause dur="0.4"/> resulting demand <trunc>cu</trunc> or or residual demand curve </u> <pause dur="0.2"/> <u who="sf0756" trans="pause"> so it's the dominant curve </u> <pause dur="0.4"/> <u who="nm0754" trans="pause"> so it's the <trunc>des</trunc> dominant firm # residual demand curve <pause dur="0.3"/> it then chooses its its optimum price <pause dur="1.1"/> and the fringe will then fill if you like fill the gap </u> <u who="sf0756" trans="overlap"> rest </u> <u who="nm0754" trans="latching"> yes <pause dur="1.5"/> # <pause dur="1.5"/> now <pause dur="2.5"/> some observations on that model <pause dur="3.1"/> if you like juxtaposing <pause dur="0.5"/> some fairly casual empirical observations with the theoretical model <pause dur="0.9"/> one of those is this that it's been observed over quite long periods of time <pause dur="0.6"/> that dominant firms once they have achieved a position of dominance tend to be around for a very long while <pause dur="1.2"/> they <pause dur="0.4"/> their market share may decline <pause dur="0.9"/> # over time and i'm now talking about very often decades rather than just a <trunc>sh</trunc> few months or <pause dur="0.3"/> even a couple of years <pause dur="0.7"/> their market share may decline

over quite a long period of time <pause dur="0.3"/> but on the whole there are some <pause dur="0.2"/> spectacular exceptions but on the whole <pause dur="1.0"/> dominant firms once they've achieved a position <pause dur="0.8"/> <trunc>a</trunc> as i've been talking about in their market <pause dur="1.2"/> they don't give up their dominance very easily <pause dur="0.8"/> some firms <pause dur="0.2"/> # in <trunc>w</trunc> # in our <pause dur="0.3"/> theoretical terms # in our fringe <pause dur="0.3"/> may encroach <pause dur="0.3"/> on the be able to encroach on the position of the dominant firm <pause dur="0.5"/> but their ability to do that <trunc>ap</trunc> apparently from the historical records seems to be quite <pause dur="0.5"/> # limited <pause dur="1.8"/> in addition to that we must reckon with the fact that the dominant firm may have as long as it can get away with it from a policy perspective <pause dur="0.4"/> may have a number of strategies open to it <pause dur="0.4"/> for ensuring that <pause dur="0.3"/> even if its market share has declined for a

little while over a few years <pause dur="1.2"/> it can try to regain that <pause dur="0.6"/> # position <pause dur="0.9"/> # by using a number of strategies <pause dur="1.1"/> # the most extreme of those i suppose until quite recently <pause dur="0.4"/> was the dominant firm in the British <trunc>s</trunc> # cement industry <pause dur="0.4"/> where # <pause dur="0.3"/> for for a number of <pause dur="0.2"/> # extraneous reasons <pause dur="0.2"/> the market share of this dominant firm is known for a very long period indeed throughout <pause dur="0.5"/> # the nineteen-hundreds <pause dur="0.6"/> # sorry throughout the from nineteen-hundred to the year <pause dur="0.5"/> about nineteen-eighty nineteen-ninety <pause dur="0.5"/> its market share <pause dur="0.9"/> almost remained unchanged what happened was that this dominant firm <pause dur="0.6"/> achieved its dominance through # a massive merger in just prior to the First World War <pause dur="0.9"/> it had # a very large market share then it started declining <pause dur="0.2"/> somewhat <pause dur="0.2"/> what it did was to <pause dur="0.2"/> to <trunc>o</trunc> over a number of decades it simply <pause dur="0.2"/> when that happened it simply acquired its nearest rival <pause dur="1.4"/> now you could say well that is now that sort of behaviour is now constrained <pause dur="0.4"/> by # competition policy and that would be <pause dur="0.2"/> correct <pause dur="0.7"/> but it seems to have

been able to get away with that strategy for a quite long while so even though the <pause dur="0.4"/> the fringe if you like <pause dur="0.2"/> # in that market was <pause dur="0.8"/> the the the <trunc>w</trunc> the the amount to this <pause dur="0.5"/> # fringe supply curve <pause dur="0.4"/> getting more elastic the fringe would be supplying more and more <pause dur="0.5"/> to recover its position <pause dur="0.3"/> in that example the firm simply acquired <pause dur="0.5"/> its nearest rival <pause dur="0.3"/> so one way of retrieving <pause dur="0.2"/> its market share was by acquisition <pause dur="1.8"/> # <vocal desc="cough" dur="1" iterated="y"/> another way of shifting this curve <pause dur="0.2"/> to the dominant firm's advantage <pause dur="0.3"/> may be for example <pause dur="0.2"/> if the dominant firm <pause dur="1.2"/> in a sense overbuys a crucial input <pause dur="0.2"/> both the dominant firm <pause dur="0.8"/> and <pause dur="0.3"/> the fringe firms may have to <trunc>sup</trunc> <pause dur="0.3"/> have to use certain crucial inputs <pause dur="1.1"/> if the dominant firm deliberately as a piece of strategy <pause dur="1.0"/> buys up in any a particular time period more than it knows it's going to use for its own output <pause dur="0.4"/> that will tend to push up the price of that input <pause dur="0.5"/> and raise the cost not only to the dominant firm <pause dur="0.5"/> but to the fringe firm <pause dur="0.3"/> and you may say well <pause dur="0.7"/> why push up its own costs <pause dur="1.2"/> <trunc>i</trunc> i know

that the fringe firm's costs are # rising as well but why do that surely as a piece of strategy it seems very shortsighted <pause dur="0.3"/> but as far as the dominant firm is concerned <pause dur="0.5"/> it can do a rough calculation <pause dur="0.6"/> that if it can impose on the <trunc>admitted</trunc> already higher cost fringe firms <pause dur="0.5"/> greater burdens than itself <pause dur="0.2"/> 'cause as the dominant firm it may be more capital intensive <pause dur="0.5"/> and these may be <pause dur="0.5"/> # <pause dur="0.3"/> variable costs rather than fixed costs <pause dur="0.2"/> so long as the <pause dur="0.3"/> amount of cost increased that it can impose <pause dur="0.2"/> on the fringe <pause dur="0.3"/> are greater <pause dur="1.2"/> than on itself <pause dur="0.5"/> and if it does retrieve <pause dur="0.3"/> its market share <pause dur="0.2"/> in so doing that is it therefore generates more revenue <pause dur="1.1"/> it will be worthwhile it doing that <pause dur="0.3"/>

so one way of shifting <pause dur="1.0"/> the fringe supply to its own advantage <pause dur="0.3"/> would be to overbuy <pause dur="0.2"/> on a crucial input <pause dur="1.3"/> thus pushing up the price both to itself <pause dur="0.3"/> and to <pause dur="0.4"/> # its fringe rivals but so long as its retrieval of its market share <pause dur="0.3"/> generates an <trunc>a</trunc> increase in revenue greater than the increase in cost for it <pause dur="0.4"/> it will be worth its while <pause dur="1.7"/> another more # <pause dur="0.3"/> the same sort of strategy might be for the dominant firm to try to press <pause dur="0.3"/> through lobbying <pause dur="0.5"/> for <pause dur="0.5"/> a change in the regulations for example governing the industry <pause dur="1.0"/> which may have a greater impact on the smaller fringe firms than on it as the established dominant <pause dur="0.2"/> firm <pause dur="1.4"/> there are a number of ways <pause dur="0.2"/> that it might be able to do that <pause dur="0.2"/> for example if the change regulation affects <pause dur="0.3"/> only new entrants <pause dur="0.3"/> rather than established firms <pause dur="0.2"/> their costs <pause dur="0.3"/> may be <pause dur="0.2"/> higher <pause dur="0.6"/> or the increase in cost may be

higher <pause dur="0.2"/> than the increase in costs for the dominant firm <pause dur="2.6"/> a number of strategies might be possible therefore for changing <pause dur="0.4"/> the <pause dur="0.5"/> # the position of this <kinesic desc="indicates point on transparency" iterated="n"/> fringe supply <pause dur="0.2"/> if it's successful what that would mean <pause dur="0.2"/> is that it would be moving up in that direction it'd be rising <pause dur="0.7"/> so that at each price <pause dur="0.2"/> the fringe will be able <pause dur="0.2"/> to supply <pause dur="0.3"/> less <pause dur="0.8"/> if the dominant firm is successful <pause dur="0.7"/> # <pause dur="0.7"/> in raising <pause dur="0.9"/> both its own <pause dur="0.3"/> and <pause dur="0.2"/> costs to the fringe <pause dur="1.8"/> as another piece of strategy it might attempt to shift the residual demand curve we <trunc>talk</trunc> i talked about <pause dur="0.3"/> shifting <pause dur="0.5"/> the fringe supply curve <pause dur="1.2"/> what it might also try to do is to <pause dur="0.2"/> shift <pause dur="0.2"/> the residual demand curve <pause dur="2.4"/><kinesic desc="indicates point on transparency" iterated="n"/> it might do this and this will depend of <trunc>cou</trunc> to a <trunc>s</trunc> # a large extent on the nature of the product <pause dur="0.2"/> it might do this by undertaking for example a very heavy <pause dur="1.0"/> # advertising campaign to swing demand towards its product away from the products of the fringe supply <pause dur="0.7"/> now that what that would do <pause dur="0.2"/> would be to push out the <pause dur="0.4"/> residual <pause dur="0.2"/> demand curve <pause dur="0.3"/> meaning that it

would <trunc>comm</trunc> not only command a higher price <pause dur="1.5"/> but the dominant firm would supply more of the market <pause dur="0.6"/> than <pause dur="0.2"/> previously <pause dur="1.8"/> and again <pause dur="0.3"/> as long as the <pause dur="1.0"/> shift in the demand which would generate more revenue for the firm the dominant firm so long as that is greater <pause dur="0.7"/> than the increasing cost that it's incurring through in my example the advertising <pause dur="0.3"/> it would pay it <pause dur="0.3"/> to <pause dur="0.4"/> # to do that <pause dur="2.5"/> as an alternative <pause dur="0.3"/> to <pause dur="0.2"/> that sort of strategy <pause dur="0.4"/> that is in in shifting <pause dur="0.6"/> directly shifting the residual demand or the fringe supply <pause dur="0.4"/> much more aggressively <pause dur="0.3"/> the dominant firm and i think i'm # i <pause dur="0.7"/> hinted at this <trunc>a</trunc> in my one of my earlier lectures <pause dur="0.5"/> the dominant firm may actually <pause dur="2.8"/> decide to <pause dur="0.7"/> try to exploit much more aggressively the learning curve you remember i mentioned that <pause dur="0.5"/> in some industries not by no means all but in some industries learning effects are very very important <pause dur="1.5"/> the dominant firm may therefore try to move very very rapidly down that learning curve which implies that it would charge a very low price <pause dur="1.1"/> now

in the limit in our <pause dur="0.5"/> theoretical scheme of of the such a market <pause dur="0.6"/> it may actually push the price towards or at or near P-<pause dur="0.2"/>0 <pause dur="0.6"/> in which case <pause dur="0.4"/> the amount of fringe supply would be very very small <pause dur="1.6"/> note that by assumption i don't think it's a <pause dur="0.6"/> # it's not a very <trunc>imp</trunc> i think it's very plausible assumption <pause dur="0.3"/> the assumption however is that the dominant firm has lower costs to start with <pause dur="1.5"/> if <pause dur="0.3"/> and we'd have to change <pause dur="0.5"/> in my learning curve case <pause dur="0.3"/> we'd probably have to <trunc>ma</trunc> <pause dur="0.3"/> draw a different sort of marginal cost curve <pause dur="0.7"/> but if there are important learning effects <pause dur="0.3"/> the dominant firm may be able <trunc>t</trunc> or may <pause dur="0.7"/> # <trunc>mo</trunc> or try to move very rapidly down that <pause dur="0.3"/> learning curve <pause dur="0.2"/> charge a very low price <pause dur="0.4"/> which means that the fringe supply <pause dur="0.6"/> # would be the {st}fringe suppliers would be struggling <pause dur="0.3"/> and their <pause dur="0.5"/> # <pause dur="0.8"/> market share <pause dur="0.2"/> would be reduced <pause dur="1.9"/> important <pause dur="0.2"/> questions i will leave you with because i want to take it up bit later on in a later lecture is this <pause dur="0.5"/> if the dominant firm did see it is it in its own interests as <pause dur="0.3"/>

as charging a price below P-zero <pause dur="0.4"/> or P-O in this diagram <pause dur="0.9"/> if it decided that it would be in in its own interests in the short run to price below P-<pause dur="0.3"/>O <pause dur="0.4"/> is that predatory <pause dur="1.3"/> is it <pause dur="0.4"/> anti-competitive <pause dur="2.2"/> and i will take up the question <pause dur="0.4"/> of # <pause dur="2.2"/> predatory behaviour in a subsequent lecture in more detail and try to <pause dur="0.5"/> address <pause dur="0.2"/> that question which for the moment <pause dur="0.4"/> i will leave you with <pause dur="0.7"/> let me sum up then on the <pause dur="0.4"/> # outcome of a model of this kind <pause dur="1.8"/> the prices <pause dur="0.3"/> if you have a structural of of market where you have a dominant firm and a fringe supply the prices <pause dur="2.5"/> with the fringe are lower <vocal desc="cough" iterated="n"/> than without the fringe <pause dur="0.2"/> and of course the output is therefore higher <pause dur="0.4"/> consumers therefore benefit <pause dur="9.6"/> historically <pause dur="0.7"/> it's the second point historically <pause dur="0.6"/> dominant firms' shares have tended to erode but that erosion <pause dur="0.2"/> is very slow <pause dur="0.5"/> generally <pause dur="2.7"/> even if they can't <pause dur="0.3"/> use <trunc>mer</trunc> the merger route to reclaim their or regain their market share <pause dur="2.7"/> because of competition policy <pause dur="0.6"/> they may be able to recover <pause dur="0.2"/> their market share <pause dur="0.3"/> more <pause dur="0.4"/> # <pause dur="0.2"/> subtly if

you like <pause dur="0.5"/> by <pause dur="0.3"/> trying to shift either the fringe supply curve <pause dur="0.3"/> or the residual demand curve <pause dur="2.6"/> my final point on that is that in the same study which i referred to earlier on this morning <pause dur="0.8"/> # <pause dur="0.2"/> over a quite a long period it varied a little bit but it was something of the order of forty or fifty years in a sample of nineteen <pause dur="0.4"/> dominant firms defined in the way that i <trunc>defi</trunc> <pause dur="0.3"/> i mentioned earlier on <pause dur="0.5"/> that is the dominant firm had fifty plus <pause dur="0.3"/> share of the market <pause dur="1.6"/> thirteen of those shares declined in this thirty to forty year period <pause dur="1.0"/> or remained unchanged <pause dur="0.2"/> <vocal desc="cough" iterated="n"/> within a <pause dur="0.2"/> percentage point or two thirteen of <pause dur="0.2"/> a sample of <pause dur="0.4"/> nineteen <pause dur="1.2"/> declined or remained unchanged <pause dur="0.9"/> and six increased <pause dur="0.4"/> in this period <pause dur="0.2"/> so these they were already dominant <pause dur="0.3"/> they increased <pause dur="0.2"/> their share <pause dur="1.5"/> the final point <vocal desc="cough" iterated="n"/> i will make on this <pause dur="0.4"/> # from that same study is that <pause dur="0.5"/> in each case even though they declined <pause dur="0.5"/> they remained the market leader <pause dur="1.3"/> so i will leave you <pause dur="0.2"/> with <pause dur="0.2"/> two things <pause dur="1.2"/> dominant firms take a long while to decline but they <trunc>re</trunc> <pause dur="0.2"/> tend to <trunc>re</trunc>

even if they do decline they retain their dominance or their <pause dur="0.3"/> leading <trunc>p</trunc> position in the market <pause dur="1.3"/> secondly <pause dur="1.1"/> which i will open with # next Friday second question <pause dur="1.3"/> if dominant firms remain <trunc>a</trunc> around a long while and apparently their market share only declines fairly gently <pause dur="1.7"/> in addition to the strategies that i have already mentioned are there other strategies that they can use <pause dur="0.4"/> to <pause dur="0.2"/> prevent <pause dur="0.2"/> their share from eroding from new entry <pause dur="0.6"/> and the question i will address at the beginning of next time <pause dur="0.2"/> is precisely that <pause dur="0.4"/> is it <pause dur="0.2"/> logical for example for dominant firms <pause dur="0.2"/> to use limit pricing that is <pause dur="0.5"/> to charge a lower price in the short run <pause dur="0.6"/> to deter entry <pause dur="1.4"/> in order to maintain a higher profits <pause dur="0.4"/> for themselves in the long run <pause dur="0.9"/> and i will stop there