Tied Pubs Final Report

Modellingtheimpactofproposedpoliciesonpubs
andthepubsector
AreporttotheDepartmentforBusiness,InnovationandSkills
ProjectReference:008/1314
Preparedby
December2013
AboutLondonEconomics
LondonEconomicsisoneofEurope'sleadingspecialisteconomicsandpolicyconsultanciesandhasitshead
office in London. We also have offices in Brussels, Dublin, and Cardiff, and associated offices in Paris and
Valletta.
We advise clients in both the public and private sectors on economic and financial analysis, policy
developmentandevaluation,businessstrategy,andregulatoryandcompetitionpolicy.Ourconsultantsare
highlyqualifiedeconomistswithexperienceinapplyingawidevarietyofanalyticaltechniquestoassistour
work, including costbenefit analysis, multicriteria analysis, policy simulation, scenario building, statistical
analysis and mathematical modelling. We are also experienced in using a wide range of data collection
techniquesincludingliteraturereviews,surveyquestionnaires,interviewsandfocusgroups.
HeadOffice:7175SheltonStreet,CoventGarden,London,WC2H9JQ,UnitedKingdom.
w:www.londecon.co.uk e:info@londecon.co.uk
t:+44(0)2078668185
f:+44(0)2078668186
Acknowledgements
LondonEconomicswouldliketothankthepubcompanieswhosupplieduswithconfidentialdata,Compass
LexeconandFDIConsultingwhohelpedustoaccessthedata,KateNichollsoftheAssociationofLicensed
MultipleRetailersandhermembershipwhoprovidedinsightsintokeyassumptions,AndyTigheattheBBPA
who provided key assistance, and Phil Dixon from PICAS / PIRRS whose views were both informative and
veryenlightening.Allerrorsare,ofcourse,thesoleresponsibilityofLondonEconomics.
Authors
RichardHeys,GavanConlon,MargueritaLane,andMaikeHalterbeck.
WhereverpossibleLondonEconomicsusespapersourcedfromsustainablymanagedforestsusingproductionprocesses
that meet the EU ecolabel requirements. Copyright 2013 London Economics. Except for the quotation of short
passagesforthepurposesofcriticismorreview,nopartofthisdocumentmaybereproducedwithoutpermission.
Contents
Page
Executivesummary
iii
1 Introduction
1.1
1.2
Context:theBritishpubindustry
Objectives
1
3
2 Overviewoftheproposedpolicyoptions
2.1
2.2
PolicyOptionI:AStatutoryCodeofPracticeandAdjudicator
PolicyOptionII:Amandatoryfreeoftieoption
4
8
3 Researchmethodology
3.1
3.2
3.3
3.4
3.5
TheBISapproach
TheLondonEconomicsapproach
Datacollection
Modelsetup
Employmenteffectsofclosure
9
13
15
18
20
4 Resultsandconclusions
4.1
4.2
4.3
22
Whymightapubclose?
Modelledoptionsandresults
Conclusions
22
22
32
5 References
34
LondonEconomics
Modellingtheimpactofproposedpoliciesonpubsandthepubsector
Tables,Figures&boxes
Page
Table1:
Theimpactonpubclosuresandemployment(firstordereffectsonly)
vi
Table2:
Theimpactonpubclosuresandemployment(secondroundeffects)
vi
Table3:
Examplebalancesheet,basedonBISsamplerentassessmentstatement
12
Table4:
Numberoftiedpubsbecomingunviableanddivestedbypubcos(firstordereffects)
underdifferentscenarios
23
Table5:
TotalnumberoftiedPubsclosingasaresultofthepolicy(firstordereffects)
24
Table6:
Joblossesfrompubclosures(firstordereffects)
25
Table7:
Theimpactonpubclosuresandemployment(firstordereffectsonly)
26
Table8:
Tiedpubsbecomingunviableanddivestedbypubcos(secondordereffects)
28
Table9:
Totalnumberoftiedpubsclosing(secondordereffects)
29
Table10: Joblossesasaresultofpubclosures(secondordereffects)
30
Table11: Theimpactonpubclosuresandemploy(secondroundeffects)
31
Figure1:
ThecompositionoftheBritishpubindustry
ii
LondonEconomics
Executivesummary
Executivesummary
Introduction
ThisreporthasbeendeliveredbyLondonEconomicstotheDepartmentforBusinessInnovation
and Skills on the basis of commercially confidential data that the pub companies (pubcos)
generouslypermittedLondonEconomicstohaveaccesstoforthepurposeofthisstudy.Thisdata
isdesignedforcorporatefinancialpurposes,notnecessarilyforresearchpurposes.Inanumberof
caseswehavehadtotakekeyaggregatedataanddisaggregatethis using thebestassumptions
availabletodeliverapubbypubassessmentofthetiedpubsector.
This approach delivers a model which we have calibrated to the existing state of the market,
specificallytheexistingrateofclosures,butwhichisthereforenomorethanourbestachievable
estimateofreality.Aswithanygenerateddatatheassumptionsusedinallocatingcoststopubs
couldbeverypowerfulindeterminingtheoutcomeoftheanalysis.
Assuch,whilstwebelievetheconclusionswedrawbelowpresentasignificantstepforwardinthe
debate in terms of explaining in some depth the market dynamics faced both by pubcos and
individualpubs,intermsofthepotentialimpactofthereformsproposedintheconsultationwe
feelitisnecessarytointroduceameasureofcaution.
Thetiedpubsystemisoneofthemostinterwovenindustrialrelationshipsyoucanidentifyinthe
UK, with multiple streams of payments running in both directions, from the pub tenant to the
pubco and vice versa, generally negotiated on a pubbypub basis. Unravelling these streams,
including'specialcommercialorfinancialadvantages'(SCORFA)which,bycommon(dis)agreement
are extremely difficult to quantify, has forced us to develop an approach based on testing key
assumptionsthroughrunningmultiplescenariostoproducearangeofoutputsaswedonotfeel
thatanymodelbasedonthecurrentstateofknowledgecouldbetrustedtodeliveravalidpoint
estimate 1 ,especiallyinthelightoffundamentalunderlyingrealignmentsindrinkingbehaviourin
theUK.
Inshort,theseestimatesshouldbeusedtoinformthepolicydebate,andtoprovideinsightinto
thefinancialpositionsofpubsandpubcos,butcannotbetakentoprovideacastironestimateof
potentialimpactsofpolicyreform.Oneofourkeyassumptionsisthetransferrequiredtodeliver
thenoworseoffcondition.Thiscannotbeestimateddirectlysoinsteadwehavemodelledfour
scenarios that represent the range of feasible possible values. Importantly, this means the
scenarios presented in this report estimate the full possible range of impacts rather than being
directestimatesofwhatwillhappen.
Apointestimateisasinglenumberoutputfromamodel,suchasmymodelpredictsateamwillscoreonegoalasopposedtoa
rangeestimate, such as my model predicts a team will score between zero and three goals. A pointestimate reflects far higher
certaintyinthemodellingthanwefeelwecandeployinthiscase.
LondonEconomics
iii
Executivesummary
Thetiedpubmodel
AlthoughpubsconstituteanimportantpartofBritainssmallbusinessnetworkandarebasedona
longstandingtradition,theBritishpubsectorhasfacedconsiderablehardshipthroughoutrecent
decades. Some of the problems originate in shortterm factors such as the ongoing economic
recession,buttherearealsolongertermfactorsatplay,suchastheintroductionofthesmoking
ban, competition from supermarkets, an increased tendency towards alcohol consumption at
home,aswellasothersocialanddemographicchanges 2 .Asaresultofthesedevelopments,the
pubindustryhaswitnessedasignificantcontraction,fromalmost70,000pubsin1980to51,000
today,with18(net)pubsclosingeveryweekoverthelastsixmonths 3 .
NearlyhalfofpubsintheUKaretiedpubs;ahistoricifunusualbusinessmodelwherethetenant
landlordpaysalowerpropertyrent(dryrent)totheirpubcompany(pubco)landlord,andreceives
'special commercial or financial advantages' (SCORFA), such as free satellite television or
subsidised buildings insurance. In return for this, they commit to only purchase beer from the
pubco,oftenatabovemarketprices(wherethesurplusislabelledthewetrent).
Reviews by the OFT 4 have identified that competition in the pub market is sufficient to ensure
consumersarenotdisadvantaged,andthatthetiedpubsystembringsadvantages,particularlyin
contributing to improving the production or distribution of goods or in terms of promoting
technical or economic progress, while allowing consumers a fair share of the resulting benefit
through:
Themaintenanceofavibrantanddiversedomesticbrewerysector,particularlyrelatedto
nonlager products, through guaranteed sales leading to economies of scale in
production,
Themaintenanceofsmall,wetled 5 pubswhichformacommunityhubandprovidesocial
cohesion,and,
Themaintenanceofalargernumberofpubstocompetewitheachother.
The proposed reforms aim to deliver a healthy pubs sector, where pub companies treat their
tenants fairly, where fairly is defined in that tied tenants are no worse off than freeoftie
tenants,asmeasuredbytheshareofprofitstheyareabletoretain.
Ourobjectives
Thisstudyhasthefollowingobjective:
Toprovideanindependentanalysisoftheimpactoftheconsultationproposalsongross
and net pub closures and employment levels within the pub industry, using whatever
robust evidence can be accessed, to inform the final stage impact assessment of the
likely costs and benefits of the proposals. In particular, the analysis should model the
OfficeofFairTrading,2010
DepartmentforBusiness,InnovationandSkills,2013a
4
NotingtheOFTsfindingthatbecausethepubsectorwascompetitivethismeantthatconsumerswerenotfacingartificiallyinflated
prices(OfficeofFairTrading,2010,sections2.312.33).
5
Apubwhichreceivesthemajorityofitsincomefromthesaleofalcoholicandcarbonatedbeveragesiswetled.Onewhichreceives
mostofitsincomefromfoodsalesisdryled.
3
iv
LondonEconomics
Executivesummary
potential impact of a mandatory freeoftie option and the impact of introducing a

StatutoryCodeonpubclosuresandemploymentlevels.
Todeliverthis,theDepartmenthasaskedustosetoutaneconomicmodelofthepubindustry
whichfacilitatestheestimationofthemainimpactsofthereformsonthecostsandrevenuesof
pubsandpubcompanies,toallowustoestimate:
Thenumberofpubswhichmightclose,and
Thepotentialimpactonemploymentlevels.
Finally there is a requirement to test the sensitivity of modelling results to changes in key
assumptions.
Resultsandconclusions
Theproposalsunderdiscussionaregenerallycharacterisedassimplyreallocatingprofitfromthe
pubcompany(pubco)tothetenant,soifwearedoingnothingmorethanredirectingprofitfrom
oneagenttoanother,whyshouldthischangetheviabilityofthepub?
The answer is that the discussion does not revolve around profit but rather around operating
margin,whichisadifferentconcept.
Profit is the residual from revenue after all costs have been covered. Operating margin is the
revenueremainingafterfrontlineoperationalcostshavebeenpaid.Thedifferenceisthatwhilsta
pubmaybeviablefromthetenantsperspectiveintermsoftherevenueexceedingtheoperating
costs and the cost of sale goods the tenant faces, there are further costs, which are currently
bornebythepubcoandthatmustalsobepaidfor.
The most significant of these are any capital costs (e.g. maintenance), which are carried by the
pubco,anddebtservicingcosts(i.e.theinterestonthedebtusedtopurchasetheproperty).The
basic argument here is directly paralleled by the fact that many private households finances
wouldlooksignificantlybetteriftheywerenotpayingtheinterestontheirmortgage,orpayingall
thecostsofrepairsandimprovementstotheproperty.
For the pubco, the position is relatively simple. Income from the pub needs to be sufficient to
coverboththesecosts,andthemanagementcostswhichthepubcofacesaroundtheseelements
and the contract with the tenant, as well as covering the cost of any SCORFA provisions. If the
incomefromthepubfallsbelowtheaggregateofthese,thepubmaycontinuetoappearviableto
thetenantbutwouldnolongerbeviableforthepubco,whichwouldbeforcedtodivestthisasset.
Wehavemodelledfiveoptionstoreflecttheproposalsputforwardintheconsultation:
Thecurrentbasecase
ImplementingtheCode,butnotallowingaguestbeeroption
ImplementingtheCodewiththeguestbeeroption
Allowingtenantstogofreeoftie,withalltenantstakingthisoption
LondonEconomics
Executivesummary
Allowingtenantstogofreeoftie,butonlydoingsoifitimprovedtheirfinancialposition 6 .
Wehavemodelledtheconsultationproposalsusing36scenarios,basedonthekeyvariablesofthe
value of SCORFA, the divisible balance percentage and the dryrent adjustment. The results
presentedprovidetheimpactofthealternativescenariosrelativetothecurrentbaselinescenario.
Table1: Theimpactonpubclosuresandemployment(firstordereffectsonly)
Minmaxranges
Percentageofallpubsclosing
Percentageoftiedpubsinscope
ofcodeclosing
Numberofpubsclosing
Numberofjobslost
Option1no
guestbeer
Option1
guestbeer
Option2
freeoftie(all)
Option2
freeoftie
(optimised)
514%
1123%
1419%
1420%
1136%
2858%
3548%
3550%
1,5004,800
7,50023,700
3,8007,700
18,80038,100
4,6006,400
23,00031,600
4,6006,700
23,00033,100
However,thistableonlyaccountsforthefirstordereffects,orthedirecteffectsofthechange.
Withthisnumberofpubspotentiallyclosing,weneedtoalsoconsiderthesecondordereffects,
or knockon implications of the change. The most important of these is the behaviour of
consumerswhopreviouslywouldhaveusedaclosedpubandwhonowhaveachoice:
Stopdrinking
Stopusingapubanddrinkatadifferentvenue(home,club,baretc)
Visitadifferentpub
Theliteratureonpubswitchingbehaviourisverythin.Wehaveassumedthat60%ofcustomers
whose pub shuts move to another pub. We have added in this residual demand (and therefore
revenues)tothepubsmodelledtoseewhatimpactwecouldexpectthistohave.Asshownbelow,
thisactstobringtherangeofpubsestimatedasclosingdownwards.
Table2: Theimpactonpubclosuresandemployment(secondroundeffects)
Minmaxranges
Percentageofallpubsclosing
Percentageoftiedpubsinscope
ofcodeclosing
Numberofpubsclosing
Numberofjobslost
Option1no
guestbeer
Option1
guestbeer
Option2free
oftie(all)
Option2free
oftie
(optimised)
01.4%
0.93.3%
1.52.6%
1.52.8%
05.3%
3.212.3%
5.79.9%
5.510.7%
0700
03,500
4001,600
2,1008,100
7001,300
3,7006,500
7001,400
3,7007,000
ThevastmajorityoffreeoftiepubsareresponsibleformaintenanceanddonotreceiveSCORFA,whichmeanstheadditionalincome
frombeingfreeofthewetrentneedstoexceedthesumoftheseforatenanttowishtousetheoption.
vi
LondonEconomics
Executivesummary
Finally,anothertypeofsecondordereffectwhichcouldoccuristhatpubcoscouldreducetheir
managementcostsandmaintenancebudgetsandrefinancetheirdebttoreducetheircostsinthe
faceofthisreform.Weanticipatetheywouldattempttodothis,butgiventhedegreetowhich
the pubcos have faced losses in recent years, we believe they may now find it difficult to go
significantlydeeperintermsofefficiencies.Assuch,wehavenotmodelledthiseffect.Inpartwe
havenotattemptedtoestimatethiseffectbecausebothclosureandpubsturningfreeoftiecould
reducetheeconomiesofscalethatpubcoscurrentlyachieve,drivinginadditionalcostpressures
onthosetiedpubswhichremain.
Conclusions
ThetiedpubmodelisoneofthemostcomplexindustrialrelationshipsremaininginUKindustry
and aims to deliver a system to ensure floor levels of demand for British brewers, sustaining
diversityandthetraditionalfamilybrewerymodel.Inrecenttimes,thishasbecomeincreasingly
necessary, as the consumption of beer in pubs has declined, with the numbers of barrels sold
falling dramatically as consumption patterns have changed, both due to lower prices in
supermarketsandtheemergenceofalternativessuchasrestaurants,clubs,andbars.Anumberof
stakeholders interviewed noted that the UK is probably still operating excess pub supply of
approximately6,000pubs,suggestingasustainablenumberofpubsofapproximately45,000.
Thisoversupplyhasled tolowprofitability,bothformanytenantsandpubcos,particularlyina
climatewhereservicingdebthasbecomedifficult.Thekeyfindingofthisstudyisnotthenumber
ofpubswhichmaycloseasaresultofonepolicyoranother,butratherthehighnumberofpubs
that currently appear to be at the margin of viability. Irrespective of what changes may be
proposed or considered, the interlocking nature of a large variety of revenuestreams, and the
high level of costs being faced by pubcos, suggest that almost any policy reform may have
noticeableandunpredictableeffects.
Intheestimatesweproduceoftheimpactoftheconsultationproposals,eventakingaccountof
secondordereffects,ofthe13,300pubswebelievewillbeinscopeoftheCode,upto2,400or18
percentcouldbecomeunviablefortheirpubcoowners,ontopofthosealreadyunviable(c.1,300) 7
within the base case scenario, although we estimate a third of these would reopen under
alternativemanagement.
The threat that the number of tied pubs currently operated by the pubcos may diminish by
between 2530% due to closures, even if these pubs subsequently reopen under a different
owner,anddisregardinganywhomaywishtotakeupthefreeoftieoption,maybesufficientto
eliminatetheeconomiesofscaleinpurchasingthatmanytiedtenantsinthissector(unknowingly)
benefitfrom.Thissuggeststhatthereisarealpossibilitythateachoftheproposedpolicyreforms,
exceptpossiblythecodewithoutpermittingguestbeer,insteadofdeliveringthepolicyobjective
ofensuringtiedtenantsaretreatedfairly,i.e,noworseoffthanfreeoftietenants,mayleadto
theendofalargescaletiedpubsystem.
This,however,maynotbeasdisastrousasitinitiallysounds.Thetiedpubmodelisverysimilarto
more standard franchising arrangement, albeit one where the pubco supplies the sales product
Itmustbenotedthatthisfigureisdrivenbyourassumptionsinthecalibrationexerciseandisthereforeentirelyanartefactofour
modellingprocess.
LondonEconomics
vii
Executivesummary
and leases the property to the franchisee, and this model is already being attempted by some
pubcosandmaybeworthyoffurtherconsideration.
Intheshortterm,itisalsoworthnotingthatthePICAS/PIRRSsystem,whichisthemostconcrete
element of the current voluntary framework, is now becoming an established part of the
infrastructure, and appears to be having some effect in starting to address the worst tenant
circumstances 8 ,althoughsomestakeholdershaveraisedthepointwithusthatsometenantsare
wary of taking the PICAS/PIRRS route in case the discretion available ends up increasing their
rental,eventhoughitisrecognisedintheBISCommitteereportthatthereareflawsinthecurrent
systemtoo.
It is our conclusion that the reforms proposed in the consultation will close up to 1,600 pubs,
although there is very great uncertainty about the precise value; In particular, the size of the
transfer from pubcos to tenants resulting from the no worse off principle is very hard to
estimate, our results reflect the impact of a range of possible transfer values; However, there is
clearlysurpluspubcapacity,inquiteavolatilemarketwhere,withsomanypubsonthemarginof
viabilityitishardtodeterminewhichpubswillclose.Stakeholdersraisedwithusthatthatthere
maybeupto6,000surpluspubsintheUKforming12%ofthemarket,andifthispolicydiddeliver
severalthousandclosedpubsitwouldactasasubstantialfractionofthislongtermtrendwhichis
likelytooccurunlessmajorchangestotaxpolicyandsocialnormstakeplace.
If one assumes, as we have done, 60% of consumers move toanother pub that implies that, on
averagepubswhichremainwillseefootfalls7.2%higherthanpresent.Thiswouldbesufficientto
turn a poorly performing pub into a more attractive prospect if it can see the immediate future
out.Assuchitmaydeliverenoughofaboosttootherpubstoreduceclosureratesinthemedium
term.
Business,InnovationandSkillsCommittee(2011).
viii
LondonEconomics
1Introduction
Introduction
1.1
Context:theBritishpubindustry
AlthoughpubsconstituteanimportantpartofBritainssmallbusinessnetworkandarebasedona
longstandingtradition,theBritishpubsectorhasfacedconsiderablehardshipinrecentdecades.
Someoftheproblemsoriginateinshorttermfactorssuchastheongoingeconomicrecession,but
there are also longerterm factors at play, such as the introduction of the smoking ban,
competitionfromsupermarkets,anincreasedtendencytowardsalcoholconsumptionathome,as
well as other social and demographic changes 9 . As a result of these developments, the pub
industry has witnessed a significant contraction, from almost 70,000 pubs in 1980 to
approximately51,000today,with18pubs(net)closingeveryweekoverthelastsixmonths 10 .
Within the sector, due to its longevity, community role and close relationship with domestic
brewers,abespokeoperatingmodelthetiedpubhascomeintoexistence.Thetiedpubisan
entity where the tenant landlord pays a lower property rent (dryrent) to their pub company
(pubco)landlord,andalsoreceiving'specialcommercialorfinancialadvantages'(SCORFA),suchas
free satellite television or subsidised buildings insurance. In return for this, they commit to only
purchase beer from the pubco, often at above market prices (where the additional surplus is
labelledthewetrent).
Reviews by the OFT 11 have identified that competition in the pub market is sufficient to ensure
consumersarenotdisadvantaged,andthatthetiedpubsystembringsadvantages,particularlyin
contributing to improving the production or distribution of goods or in terms of promoting
technical or economic progress, while allowing consumers a fair share of the resulting benefit
through:
Themaintenanceofavibrantanddiversedomesticbrewerysector,particularlyrelatedto
nonlager products, through guaranteed sales leading to economies of scale in
production,
The maintenance of small, wetled 12 pubs which form a community hub and provide
socialcohesion,and,
Themaintenanceofalargernumberofpubstocompetewitheachother.
As Figure 1 shows, nearly half of pubs in the UK are tied, as opposed to the three main
alternatives;managedpubswhereapubcompanyplacesamemberofstaffinasamanager,ora
freeoftie tenant, where the tenant pays a commercially set property rent and is free to sell
whichever beer products he wishes, or freehold pub where the landlord owns the property and
choosesthebeertosell.
Within this landscape, the relationship between large pubowning companies and their tied
tenants constitutes an important issue for the Department for Business, Innovation and Skills
11
NotingtheOFTsfindingthatbecausethepubsectorwascompetitivethismeantthatconsumerswerenotfacingartificiallyinflated
prices(OfficeofFairTrading,2010,sections2.312.33).
12
Apubwhichreceivesthemajorityofitsincomefromthesaleofalcoholicandcarbonatedbeveragesiswetled.Onewhichreceives
mostofitsincomefromfoodsalesisdryled.
10
LondonEconomics
1Introduction
(BIS) 13 . In recent years, tied tenants have filed numerous calls, including some grievances about
thefairnessofthisrelationshiptotheBritishInstituteofInnkeeping,MembersofParliament,the
Office of Fair Trading, and the Department for Business and Innovation Select Committee.
Reportedmistreatmentsoftenantsby pub companies,suchasunannounced largerentincrease
demands, are primarily caused by an imbalance in the relationship between these two types of
market actors. In particular, pub companies are considered to have access to more information
andtohaveresourcesattheirdisposaltoaffordlegalandsurveyorfeesincaseofdisputeswith
theirtenants,grantingthemthepossibilitytoextractlargesharesoftheirtenantsprofits 14 .Itis
commonplacetohearinthedebatesourcesclaimingtheaverageincometoapubcofromitstied
pubstobeintheregionof60,000.Incontrast,surveysindicatethattheaveragetenantretains
approximately15,000ofprofit.
Figure1: ThecompositionoftheBritishpubindustry
17%
48%
36%
Managedpubs
Freehousesandfreeoftietenantedpubs
Tiedtenantedpubs
Note:Thedatadrawnodistinctionbetweenfreehousesandfreeoftietenantedpubs
Source:DepartmentforBusiness,InnovationandSkills(2013b).
While this discrepancy in resources and information exists between pub companies and their
tenants in general, BIS is particularly concerned about the potential for and repercussions of
mistreatment of tied tenants. Beer prices change more frequently than rents, thus providing
pubcos(whocompel theirtenants topurchasebeeronlyfromthepubco(thebeertie))withan
additionalrouteofpotentialabuse.Further,thetiecomplicatestherelationshipbetweentenants
and their pub companies by increasing the amount of information to consider in the tenants
decisionmaking process. Finally, recent evidence suggests that tied tenants also face a higher
probabilityoffinancialdistress,with46percentoftiedpublicansearninglessthan15,000ayear,
comparedtoonly23percentoffreeoftietenants 15 .Alloftheseconsiderationssuggesttheneed
forBIStoputparticularfocusonthefairtreatmentofpublicanswhoarerestrictedbyabeertie.
However,thisdebatehastwosides.Thepubcosprovideanumberofservicestotiedtenantsthat
maybedifficultforthetenanttoobserve,suchasdebtfinancingonthepurchaseofthefreehold
13
Managedpubsandfreehousesarenotsubjectedtothedescribedproblemsbetweenpubownerandtenant;therefore,thesetypes
ofpubsareoflesserrelevancetothisstudy.
14
15
CAMRAsurveysquotedbyBIS,2013a
LondonEconomics
1Introduction
(without which there would not be a pub to rent), as well as maintenance and support systems
(suchasthosedeliveredviaSCORFA).Theseareeitherunobservedbytenants,ormayindeedbe
entirelyunobservable(examplesofwhichincludethedifferentialininsurancecostswhen thisis
purchasedcentrallyandinbulkbythepubcoagainstthecostthetenantwouldhavetopaywere
hetopurchasethisinisolation).
Untilrecently,theDepartmentreliedonpubcostogovernthiscomplexrelationshipthemselves,
by implementing voluntary selfregulatory approaches to treat their tenants, specifically tied
publicans, more fairly. Since 2004, the Business, Innovation and Skills Select Committee and its
predecessors have published four pub company reviews 16 . The reports have scrutinised the
difficultiesregardingtherelationshipbetweenpubcompaniesandtheirtenantpublicansandurge
pubowningfirmstoimplementmeaningfulreform.BISgrantedafinalchanceforpubcompanies
tovoluntarilymakethenecessarychangesin2011.
Acallforevidenceontheperformanceofselfregulationintheindustryin2012 17 ,foundthatthe
voluntarycodeofpractice(i.e.theIndustryFrameworkCodeofPractice)hadbeenabletoinduce
somemodestimprovementstotheproblematicrelationbetweenpubcompaniesandtenants.In
addition, BIS acknowledged the beneficial role of the Pubs Independent Review Scheme (PIRRS)
anditsPubsIndependentConciliationandArbitrationService(PICAS)inprovidingtenantswitha
costeffectivesolutiontoresolvedisputesandobtainredress.However,itwasconcludedthatthe
selfregulatory approach was not sufficiently effective at relieving tenants hardship; it was felt
that the situation could not be suitably improved without a significant culture change and a co
ordinated longterm effort to inform tenants of their rights and possibilities 18 . Therefore, in
January2013,theSecretaryofStateannouncedthattheBritishpubindustrywillbesubjectedto
statutoryinterventioninstead.
1.2
Objectives
The proposed reforms aim to deliver a healthy pubs sector, where pub companies treat their
tenants fairly, where fairly is defined in that tied tenants are no worse off than freeoftie
tenants,asmeasuredbytheshareofprofitstheyareabletoretain.
Thisstudyhasthefollowingobjective:
Toprovideanindependentanalysisoftheimpactoftheconsultationproposalsongross
and net pub closures and employment levels within the pub industry, using whatever
robust evidence can be accessed, to inform the final stage impact assessment of the
likely costs and benefits of the proposals. In particular, the analysis should model the
potential impact of a mandatory freeoftie option and the impact of introducing a
StatutoryCodeonpubclosuresandemploymentlevels.
Todeliverthis,theDepartmenthasaskedustosetoutaneconomicmodelofthepubindustry
whichfacilitatestheestimationofthemainimpactsofthereformsonthecostsandrevenuesof
pubsandpubcompanies,toallowustoestimate:
Thenumberofpubswhichmightclose,and
Thepotentialimpactonemploymentlevels.
16
Trade and Industry Committee, 2004; Business and Enterprise Committee, 2009; Business, Innovation and Skills Committee, 2010;
Business,InnovationandSkillsCommittee,2011
17
DepartmentforBusiness,InnovationandSkills,2012
DepartmentforBusiness,InnovationandSkills,2013a,2013b
18
LondonEconomics
2Overviewoftheproposedpolicyoptions
Finally there is a requirement to test the sensitivity of modelling results to changes in key
assumptions.
Overviewoftheproposedpolicyoptions
InApril2013,theDepartmentinitiatedaconsultationamongmainstakeholdersintheBritishpub
industry on the policy measures available to address the perceived problematic relationship
betweenpubcompaniesandtheirtiedtenants.Thissectionwillprovideadetaileddescriptionof
thetwomainsuggestedpolicyoptions,whoseimpactconstitutesthefocusoftheanalysisathand.
2.1
PolicyOptionI:AStatutoryCodeofPracticeandAdjudicator
2.1.1 ContentsoftheStatutoryCode
The introduction of a Statutory Code, along with an independent Ombudsman to enforce it,
constitutesthefirstpolicyoptionputforwardbyBIS.TheStatutoryCodeisintendedatgoverning
the relationship between pub companies and their tenants, putting particular emphasis on the
issuesofriskandreward,whichshouldbesharedappropriatelyamongthetwopartiesconcerned
in order to ensure the fairness of their agreements. Recognising the continuous change of
conditions and requirements of the pub industry in the United Kingdom, BIS proposes that the
newCodewillbesubjectedtoperiodicreviews.
ThemostrecentversionoftheIndustryFrameworkCode(Version6),whichhasbeenestablished
for pub companies in the industry as part of its selfregulatory approach, constitutes the
envisioned basis for the new and legally binding Statutory Code of Practice. However, BIS has
identifiedtwoareasforthepotentialstrengtheningofthiscurrentFrameworkCodeofPracticeto
includetwocoreprinciples:
1. Anoverarchingprincipleoffairandlawfuldealingaimsatwarrantingthatpubcompanies
treat all of their tenants 19 fairly, i.e. that risk and reward are shared to an appropriate
extent between themselves and their lessees, as well as lawfully, implying that any
unlawfulbehaviourbythepubcompanytowardsatenantconstitutesadefactobreachof
theStatutoryCode.Thisprincipleincludestheunderstandingthatpubcompaniesshould
conduct its relationships with tenants in good faith, without duress, and without
distinguishingbetweenformalandinformalarrangements.
2. The principle that the tied tenant should be no worse off than the freeoftie tenant
relatestotiedlesseesmorespecifically,postulatingthelattermaynotbedisadvantaged
compared to those tenants who are not subjected to a tie by their pub company. BIS
recognisesthatcompliancewiththisprinciplerequiresprecisiononhowthisprovisionis
tobeappliedtorentalcalculationsoftenants,consideringthebalancebetweenwetand
dryrent as well as SCORFA. It is currently envisioned that this second overall principle
shouldimplythattheprojectedpostrentbalanceofatiedtenantshouldbeatleastequal
totheprojectedpostrentbalance,whichthesametenantwouldreceiveunderafreeof
tielease,assumingthesameleaseortenancyconditions.
19
Thisincludesbothtiedandfreeoftiepublicans.
LondonEconomics
Putting particular emphasis on the second of these core principles BIS has brought forward two
suggestions on how to achieve the equal treatment of tied tenants as compared to freeoftie
lessees.
For the Code to require that higher beer prices due to the tie must directly be
compensated for by lower rents and/or genuinely quantifiable countervailing benefits
provided through SCORFA, thus making the postrent balance of a tied tenant at least
equaltowhatheorshewouldretainunderafreeoftieagreement,allelsebeingequal.
The additional introduction of a mandatory freeoftie option, which is a more rigorous

measuretoachievefulfilmentofthesecondcoreprinciple.Thisproposalconstitutesthe
second policy option in its own right and will be outlined in more detail in the next
section.
In addition to the two major principles just outlined, BIS suggests further areas in which the
current selfregulatory Industry Framework Code should be strengthened to establish the new
StatutoryCodeofPractice.
1. ItisproposedthatthenewCodeshouldstipulatearighttorequestanOpenMarketRent
Reviewtoforalltenants,independentoftheconditionsoftheircurrentleaseagreements,
inordertoensuretheirfairtreatmentbytherespectivepubcompany.Theconditionsfor
atenanttorequesttheOpenMarketRentReviewarethatheorshehasnotbeensubject
to a similar rent assessment throughout the last five years; that their pub company
recentlysignificantlyincreasedthepriceoftheirtiedproducts;ortherecentoccurrenceof
aneventwhichwasoutsideofthelesseescontrolandwhichwasunanticipatedatthelast
rentreview,butwhichinvolvesasignificantimpactonthetenantsabilitytotrade,suchas
the closing of a large employer nearby, dramatically reducing footfall and custom in the
area.
2. The new Statutory Code should include provisions that increase the transparency of the
relationship between pub companies and their tenants, requiring pub companies to
provide potential lessees with a range of relevant information on the costs and risks of
trading before a tenancy agreement is made. Of particular relevance, pub companies
should calculate tied and freeoftie exante rent assessments, using the same set of
assumptions for both, and explicitly stating the turnover, gross profit, costs, divisible
balanceandSCORFA,thusdemonstratingtoeachpotentialtenantthatheorshewouldbe
atleastequallyaswelloffunderatiedagreementthanunderafreeoftielease.
3. The Government has, on several previous occasions, proposed to remove the gaming
machinetiecurrentlyinoperation,andreinforcesthisproposalbysuggestingtheinclusion
ofaprovisionstatingthatnoproductotherthandrinksmaybetiedbythepubcompanies
intheStatutoryCodeofPractice.Thisproposalhassincebeensupersededbychangesto
the Voluntary Code, such that we will therefore consider the impact of this in the base
caseforouranalysis.
4. BIS is considering the introduction of a guest beer option in the Code, granting tied
tenantstheoptiontopurchaseandsellonedraughtbeerfromanysupplier,withoutany
controlsorrestrictionsimposedbythepubco.
5. Finally,theinstallationofflowmonitoringequipmentbypubcompaniestocontroltheir
tenants compliance with purchasing obligations remains a subject of controversy within
the British pub industry. It is currently suggested that the Code should restrict the
LondonEconomics
potentialforuseofdatacollectedwiththehelpofthisequipment,byincludingaprovision
statingthatsuchinformationmaynotbeusedtodeterminewhetheratenantiscomplying
with purchasing obligations, or considered as evidence for the enforcement of such
obligations.
2.1.2 Thresholdofapplication
In spite of its proposals to intervene in the industry, BIS aims to comply with better regulation
principlesandminimisethepolicyimpactonthosepubcompaniesthatalreadytreattheirtenants
fairly and are complying with the Voluntary Code, but obliging those firms which exploit their
tenants to make fast and significant improvements. In accordance with this objective, it is
currently perceived that the new Statutory Code of Practice outlined above should apply to all
Britishpubcompaniesthatownmorethan500pubs.Oneofthejustificationsforthisapproach
concerns the Governments overall intention to reduce regulation and grant more influence to
market forces, therefore minimising the burden on smaller pub companies that appear to have
actedresponsibly.Inaddition,thethresholdtargetsthesuggestedpolicymeasurestothosepub
companies that have the greatest market power within the industry, with 500 pubs constituting
approximately1%ofthemarket.Finally,itisarguedthatbiggerpubowningfirmswillfinditeasier
thansmallcompaniestobearthecostsoftheircompliancewiththeStatutoryCode.Inorderto
limitpubcompaniesabilitytocircumventbeingrestrictedbytheCodebysplittingupintoagroup
ofsmallerfirms,the thresholdof500 pubswill be appliedatthetoplevelofagroupof related
companies.
ForthosecompaniesproposedtobecoveredbytheCode,thenewrulesandregulationswillthen
applytoalloftheirtenanted(i.e.nonmanaged)pubs,fortworeasons.First,thisistoensurethat
afairtreatmentisgrantedtobothtiedaswellasfreeoftiepublicans.Second,ascenariowhere
the Code only applies to tied pubs could provide pub companies with an opportunity to avoid
regulation by removing the tie and instead increasing freeoftie pubs rents beyond current
market rent values. BIS prescribes that the Statutory Code will be incorporated in all lease and
tenancy agreements between pub companies and their publicans, to warrant that the latter are
awareoftheirrights.
2.1.3 AnindependentAdjudicator
InordertoensurepubcompaniescompliancewiththeStatutoryCode,anadditionalcomponent
ofthepolicyoptionconsideredbytheDepartmentforBusiness,InnovationandSkillsconsistsof
anindependentombudsmanresponsiblefortheenforcementofthenewrulesandregulations.It
is suggested that the role of the proposed Adjudicator be based on the model of the Groceries
CodeAdjudicator,postulatingtwomainfunctionsforthepubindustrysnewombudsman:
1. Arbitrationfunction:Inthecontextofthisfunction,theAdjudicatorwillhavethepower
to deliver redress to individual tenants in their relationship with the pub companies. In
particular, the Adjudicator will have the ability to deliver Open Market Rent Reviews in
accordancewithguidancepublishedbytheRoyalInstitutionofCharteredSurveyors(2010)
and the new Statutory Code of Practice, thus establishing that rent levels are calculated
fairly and not artificially inflated. The tenant can thus approach the Adjudicator for
arbitration, provided that the respective tenant has given the pub company 21 days to
resolve the dispute to the tenants satisfaction. While this possibility is granted to any
tenantwhoisdissatisfiedwiththecalculationofhisorherrent,theCodeprovidesforthe
LondonEconomics
possibility for the Adjudicator to impose fines on those tenants whose complaints have
beenfoundtobevexatiousorentirelyunmerited.
2. Investigatory function: This second main function allows the Adjudicator to undertake
proactive investigations based on reasonable grounds of suspicion that a pub company
was acting in breach of the Statutory Code of Practice. In the course of such an
investigation,theombudsmanwillhavetheability todemandinformationfromthepub
companies, and to impose a range of sanctions on them in case that the result of the
investigationevidencesanactualbreachoftheCode.
Notethat,inadditiontothesetwomainfunctions,theAdjudicatorwillprovidegeneraladviceand
guidancetopubcompanies(toensuretheircompliancewiththeCode)andtenants(tokeepthem
informedabouttheirrights)inordertoencouragetheoveralleffectivenessoftheCode.
The investigatory function of the Adjudicator is particularly important in ensuring that pub
companiescomplywiththenewrulesandregulationssetoutintheStatutoryCode,bygivingthe
AdjudicatorthepossibilitytodetectandtacklemoresystematicbreachesoftheCodethatwould
notberesolvedbyindividualtenantsthroughthemeansofarbitration.Forexample,incasethat
the pub company does not provide preparatory training to its tenants, the latter might not be
informedabouttherightswhichtheCodegrantsthem,andwouldnotoptforarbitrationoftheir
own accord. Such a scenario would require the Adjudicators power to investigate and impose
sanctions on the respective pub company. In addition, it appears possible that a pub company
mightoverestimatetheprojectedturnoverofanindividualpubduringarentreviewinorderto
generate higher rent from the given publican. In this case, the single tenant might refrain from
approachingarbitration,sinceactualturnoverisinfluencedbyavarietyoffactorsandaconscious
artificialinflationofsalesvolumesbythepubcompanyisdifficulttodetectattheindividualpub
level.Incontrast,anAdjudicatorwiththeabovepowerstoinvestigatewouldbenefitfromamore
holisticlevelofinformation,obligingthepubcompanytosupplyrentassessmentinformationona
largenumberofpubs,andthusbeingabletodetectsystematicbreachesoftheCodemoreeasily.
ItissuggestedthatthenewAdjudicatorwouldhavearangeofsanctionsathisdisposaltopenalise
pubcompaniesthatarefoundtobeinbreachoftheCode,providingthemwithflexibilitytotailor
theirpunishmentstotheseverityofthebreach.First,theAdjudicatorwillhavethepossibilityto
advance recommendations regarding the treatment of a given tenant by the respective pub
company,whichwouldbeimposedincaseofminorbreachesofincombinationwithmoresevere
punishments.Secondly,theAdjudicatorwillbeempoweredtorequirepubcompaniesinbreachof
theCodetopublishinformation 20 abouttheoffencetothepublicthroughappropriatechannels.
Thissanctionwouldservetowarnstakeholdersinthepubindustryaboutthecommittedbreachof
the Code, and thus act as a deterrent by influencing future interactions between the respective
pubcompanyandotherplayersintheindustry.Finally,inthemostsevereorincaseofrepeated
offences,theAdjudicatorwillbeabletoimposefinancialpenaltiesonpubcompanies,whichare
consideredastrongdeterrenttotheoffendingfirmandtoothercompaniesintheindustry.
The funding needed for the introduction and operation of the independent ombudsman will be
guaranteedthroughafinancialindustrylevy,witheachofthecoveredpubcompaniescoveringa
share of the estimated total annual cost of the Adjudicator of 900,000. 21 To ensure a fair and
20
ThissanctionisalsoreferredtoasthenameandshamepoweroftheAdjudicator.
21
BIS(2013b)
LondonEconomics
effectivestructureofleviesonthepubowningfirms,eachcompanysfirstyearsharewillbebased
ontheproportionofallpubswhichtherespectivefirmowns.Incontrast,inthesecondyearand
onwards, those firms who have been detected to breach the Code more frequently will have to
payaproportionatelylargershareofthecostsoftheAdjudicator.
Finally,asameanstoensurethecontinuousaccountabilityandeffectivenessoftheAdjudicator,
he will be required to report on the investigations and arbitrations conducted and evidenced
breachesoftheStatutoryCodeofPracticeonanannualbasis.Further,theSecretaryofStatewill
implement a review of the effectiveness of and the necessity for the Code and its enforcement
authorityeverythreeyears.
2.2
PolicyOptionII:Amandatoryfreeoftieoption
The mandatory freeoftie option is defined as an option in which the tenant is subject to no
purchasingobligationsofanyformandthereforetheonlysumpaidtothepubcompanyisthedry
rent 22 . As addressed above, the additional introduction of a mandatory freeoftie option
constitutesthemorerigorousalternativeamongthetwopolicyoptionstoensurethefulfilmentof
thecoreprinciples,particularlythesecondone,tobeincludedinStatutoryCode.BISrecognises
theparticularuncertaintiesinvolvedinthispossibleremovalofthetie,whichconstitutesthebasis
of pub companies business model, and, if removed, might increase the potential for instability
within the industry. However, albeit more rigorous, it is argued that this second suggested
measure involves less complexity for tied tenants, who might find it difficult to decide whether
their pub company is in breach of the Code under the first option (i.e. whether their postrent
balance is at least equal to what they would retain under a freeoftie agreement, all else being
equal).
The mandatory freeoftie option would be included in the Statutory Code and enforced by the
Adjudicator described above. BIS (2013a) acknowledges the particular importance of the
AdjudicatorssufficientenforcementoftheCodesprinciplesunderthisoption,sinceitisexpected
thatpubcompanieswillincreasetheirtenantsrentunderthefreeoftieoptioninordertomake
up for their reduced profit levels on pubs for which the beer tie is removed. As a result, the
introductionofthissecondoptionwouldrequiretheAdjudicatortoensurethatfreeoftierents
willnotincreasebeyondfairopenmarketlevels(i.e.complywithRICSguidance).
22
DepartmentforBusiness,InnovationandSkills,2013a,p.29
LondonEconomics
3Researchmethodology
Researchmethodology
3.1
TheBISapproach
The consultation document issued by BIS provided a worked example of how to carry out the
calculationstoderiveanoworseofftiedrentcomparedtothesamepubinafreeoftieposition.
In this section we describe this process in some depth because it is worth understanding the
conceptsin detail,andwhereandwhy wehavedeviatedfromsuchanapproach,mainlydueto
dataissues.
ThissectionisnotacritiqueofBISsapproach;insteaditismerelyanattempttoclearlyelucidate
wherewehavebeenconsistentwiththeapproachasinitiallylaidoutbyBISandwherewehave
implemented changes. It is worth explicitly stating that there are a number of key assumptions
thatmustbemade,whichwehavefoundalmostimpossibletovalidateinaconsistentwayacross
therangeofstakeholders(oftensimplybecausenoauthoritativedatasetexists,hasexistedoris
subjecttoanycurrentplanstocollect).Toadegreethisdata,beingonlyrequiredforthisexercise,
would be a waste of public funding to secure, as the proposed Code itself recognises that
assessmentwillneedtobecarriedoutonapubbypubbasis,whichwouldnotnecessarilybenefit
frominformationonnationalaverages.Itisnonethelessakeyweaknessofanyandallattemptsto
analysethisquestion.
The BIS approach takes a rent assessment on a tied pub and creates a hypothetical rent
assessmentifthesamepubwerefreeoftie.
3.1.1 PubincomeintheBISModel
In the BIS approach, in both the tied and freeoftie scenarios, revenue is broken down into the
followingheadings:
Variouscategoriesofalcoholicandsoftdrinks
Food
Accommodation
Other
Netmachineincome
Gross profit to the tenant after the costs of purchasing these items is identified. These costs
includethewetrentpaidtothepubco.Theresidualisthegrossprofit.
Itshouldbenotedthatnoeffortismadetoidentifywhetheratiedtenantmovingtoafreeoftie
situation, and benefiting from a higher income (because the wetrent is removed), might pass
someofthesebenefitsontoconsumersintheformoflowerpricesbecauseofthepreviousOFT
workwhichdemonstratedtheconsumermarketwascompetitive.
LondonEconomics
3.1.2 Othercosts
The BIS approach then identifies a number of cost categories 23 , assuming these are constant
across the two scenarios. To these, in the freeoftie option, are added the value of SCORFA
payments,whichinthetiedscenariorepresentsgoodsandservicesthatarereceivedforfree.In
essence,theideaisthatthecostsofoperatingremainconstant,butthatthesecostsaresharedin
thetiedscenario,whereastheyarewhollythefreeoftietenantsresponsibilityinthisscenario.
The key issue here is how to value the SCORFA in the two scenarios, when the pubco will
demonstratesignificantlymorebuyingpowerthananindividualtenant.Ifcostistakenasaproxy
forvalue(andindeed,inreality,howdoweexpectthepubcotoprovideanassessmentofwhatit
wouldcostsomeoneelsetopurchaseequivalentserviceswithoutbenefittingfromtheeconomies
of scale of the pubco), then this implies that the value of the SCORFA added to the costs of
operatinginafreeoftiescenarioneedstobehigherthanthoseusedinthetiedscenario.Inline
witheconomiesofscalewehaveobservedinotherindustrialcontexts,weconsideritlikelythata
pubco buying services for several hundred pubs should be able to secure prices at least a third
lower than each individual pub making their own arrangements. For example, the BBPA have
quotedtousthatwheretheaveragecostofprovidingSCORFAforoneoftheirmemberstoatied
pubisaround13,000,thecosttoanindividualtenantofthesameservicesintheopenmarketis
estimatedataround20,000,whichisbroadlyinlinewiththis.
SCORFA is an incredibly contentious figure, with a wide variety of estimates from different
sources,whichprobablydifferinalargepartaccordingtowhichelementsofthepubcossupport
areincluded.Wehaveidentifiedthefollowingestimatesinourstudy:
In2009,anumberofthesmallerbrewingpubcompaniessuggestedvaluationofSCORFA
of 15,000 per pub (Independent Family Brewers of Britain submission to European
Commission DirectorateGeneral for Competition, cited in Department for Business,
InnovationandSkills,2013b)
In 2009, in a submission by pub companies to the OFT in response to CAMRAs super
complaint, an estimate of 6,0008,000 on average per pub per year (quoted in
DepartmentforBusiness,InnovationandSkills,2013b)
IntheOFTsfinaldecisiononCAMRAssupercomplaint 24 ,SCORFAwascalculatedonthe
basisofincludingitemssuchasbuyingdiscounts(foritemsotherthandrinks),business
development assistance (such as training and legal support through business
development manager), investment and centralised technical and compliance services,
including business planning, development and marketing advice, and training but
excluding costs related to building repairs (which is in line with our inclusion of these
costs in the capital expenditures). The OFT estimated that the average costs to pub
companies of providing this support is 1,500 per year 25 , although they note that this
mightbesubstantiallyhigherinindividualcases.
23
Wages,salariesandotherstaffcosts,rates,utilities,repairs,maintenanceandrenewals,buildinginsurance,marketing
andpromotioncosts,telephony,consumables,cleaning,professionalfees,bankcharges,depreciationandinterest,and
miscellaneouscosts
24
25
Seeparagraph5.154,p.123ofOfficeofFairTrading(2010).
LondonEconomics
10
3.1.3 Thedivisiblebalancecalculation
Grossprofit(revenueafterthecostofpurchasingsalesgoods)minusoperatingcostsgiveswhatis
termedinthetiedpubsectorthedivisiblebalance.Thissumrepresentstheprofitwhichcanbe
shared(divided)betweenthetenantandthepubco.
IntheBISexample,inthetiedscenariothissumislowerthanthefreeoftiescenariobecausethe
wetrent is greater than the SCORFA. In the tied scenario, the wetrent is part of the costs
removedfromrevenuetogeneratethedivisiblebalance,whereasinthefreeoftiescenarioitis
not.Similarly,inthefreeoftie,SCORFAisaddedintothecosts,whereasinthetiedscenarioitis
not.
The divisible balance is divided in two parts. In the worked example, in a 50:50 share, one part
formsthedryrent,andtheremainderformsthetenantsincome.Asaresultofthefactthatthe
divisible balance in the freeoftie scenario increases, as the pubco is no longer receiving a wet
rent, the tenants income is therefore also higher. In which case, to achieve a no worse off
situationforatiedtenant,thetiedtenantsdryrentisreducedsothatthetiedtenantsincomeis
equivalenttothatwhichwouldbeearnedinthefreeoftiesector.
3.1.4 Thedriversofclosure
In paragraph 43 of the impact assessment 26 , the Department states: No pub should become
unviableasaresultofthispolicy as itjustmoves profitfromonepartyto another.Ifapubwas
viable prior to the policy the licensee and pub owning company could reach a negotiation that
maintainedthepubsviability.Itispossiblethatthepolicycouldhaveanimpactontheviabilityof
pubowningcompanies,asopposedtoindividualpubs.
Thisstatementispartial,andthereasonforthisisthatwhilstapubmaybeoperationallyviablein
termsoftherevenueexceedingtheoperatingcostsandcostofsalegoods,therearefurthercosts
which must be considered that are currently faced by the pubco, which are not reflected in the
exampledescribedabove.
The most significant of these are any capital costs (e.g. maintenance) that are carried by the
pubco,anddebtservicingcosts(i.e.theinterestonthedebtusedtopurchasetheproperty).The
basic argument here is directly paralleled by the fact that many private households finances
wouldlooksignificantlybetteriftheywerenotpayingtheinterestontheirmortgage,orpayingall
thecostsofrepairsandimprovementstotheproperty.Inclusionofthesecostswouldmakethe
apparentfinancialpositionofthehouseholdsignificantlylessattractive.
cover these and the management and tenant contractual costs that the pubco bears, as well as
coveringthecostofanySCORFAprovisions.Iftheincomefromthepubfallsbelowtheaggregate
ofthese,thepubmaycontinuetoappearviabletothetenantbutwouldnolongerbeviablefor
thepubco,whichwouldbeforcedtodivestthisasset.
From the pubcos perspective allocating a larger share of the operating margin to the tenant is
identical to the application of a sales tax; adding an additional cost pressure which reduces the
26
DepartmentforBusiness,InnovationandSkills,2013b
LondonEconomics
11
profitwhichthepubcoreceivesfromthepubinquestion.Ifitreducesprofittoapointthatitis
negativethenthepubisnolongerviable,evengiventhesamelevelofoperatingmargin.Wecan
illustrate this in the following simple example based on the worked example in the BIS
consultation 27 :
Table3: Examplebalancesheet,basedonBISsamplerentassessmentstatement
Costs/Revenuecategory
Revenue
/profit
Costs
Salesrevenue
Salescostsexcludingwetrent
Operatingcosts
265,000
(93,500)
(85,065)
Operatingmargin
86,435
Corporateoperation&management
fees(includingSCORFA)
(14,500)
Maintenancecosts
(10,000)
Debtservicingcosts
(30,000)
Overallmargin
31,935
Tenantprofitshare
Pubcoprofitshare
28,518
3,417
Commentary
Thisisthesumsharedbetweenthe
pubcoandthetenant
Indicativeestimatebasedonaverage
ofdatasuppliedbypubcos.
Thisisthetotalprofitafterthetotal
costs
Note:LondonEconomicscalculations
Whilstitisimportanttonotethatthesebroadaverageshavebeensimplyappliedtoapreexisting
examplefromtheBISconsultation,thatisnottosaythatthispubisnecessarilyaperfectaverage
pub in terms of its income and operating costs. The worked example from the BIS Consultation,
whichproposedforthispubatransferof10,875fromthepubcotothetenanttodeliverano
worse off outcome would involve driving the pubco into a lossmaking position on this pub. As
such,theviewthatthepubcotakesxthousandpoundsoutofapubhastobemeasuredagainst
the costs they face from this pub. One should look to estimate the profit they make per pub to
determinewhetherthetenantisbeingtreatedfairly.Nevertheless,whenpubsbecomeunviable
forthepubco,itisnecessarytopresumethatatleastsomeofthosepubswillshut.
Thepubcowillnecessarilyneedtotakealongtermviewinthistypeofsituationonapubbypub
basis,takingaccountoflongtermprofitability,butalsoalternativelandusagevalues,particularly
ifapubbecomeslossmaking,andlookslikelytoremainlossmakingfortheforeseeablefuture.In
thisexample,atransferofthistypeinamediumtermtenancy(suchasweseeinthetiedsector),
would result in this pub becoming unviable for the pubco to continue to operate unless the
existingdebtservicing,maintenanceandcorporatecostsfacedwerereduced. 28
Apubbecomingunviableforthepubcohastwoeffects:
27
DepartmentforBusiness,InnovationandSkills(2013a),p.48.
Primaryamongstthese,andonaveragelargerthantheothercostscoveredbythepubcocombined,isthecostofservicingthedebt.
Whilstsomeofthepubcoswehavetalkedtoarecurrentlyintheprocessofrefinancingtheirgeneraldebtportfolio,weanticipatethat
thisislikelytomakeadifferenceofatmostafewthousandpounds
28
LondonEconomics
12
Pubcos are likely to divest themselves of a pub they do not find viable, but it does not
necessarilyshut.Itmayinsteadbetakenonbyadifferentorganisationorindividual,who
maybeable toaccessfavourabletermsonthedebtrequiredtopurchasetheasset.To
maintain consistency and comparability with previous studies, we have assumed that,
about twothirds of divested pubs end up leaving the sector. However, this sector has
been through a period of divestment for those pubs where revenues have not been
sufficient to cover total costs (including debt servicing and maintenance), whereas this
policy compels pubs to devote a greater share of their revenue to profit rather than
coveringcosts.Assuchtherehastobeanexpectationthatopenandhonestnegotiations,
based on the transparent sharing of data will lead to moderated requests for reduced
rentals by tenants in the face of an expectation of closure if they push too far in their
request. There is clearly a middleground for many pubs where profits could be more
equitably shared, although one must consider the profit made by the tenant and the
pubcointhesamelight;afterthefullcoverageofcosts.
Theshuttingofpubsforcesconsumerstomovetoadifferentvenue.Thesemayinclude
anotherpub,analternativevenue(e.g.club),orthehome.Alternatively,pubclosuremay
result in the individual stopping drinking entirely. Despite these numerous leakages of
demandfromthepubsector,itisstillnecessarytoassumethattheremainingpubsmay
see sales increase to some degree, compensating both the pubco and the tenant more
thanbefore.Wehaveattemptedtomodelthiseffect,whichwouldmitigatethenumber
ofunviableandthusclosedpubs,asaresultofthisproposedreform.
3.2
TheLondonEconomicsapproach
Our approach has to be driven by the available data. At the present time, none of the relevant
pubcos systematically collect data from their tenants on the value of sales, or the costbase of
their operation. It is therefore not possible to observe an accurate estimate of the divisible
balance.
The pubcos do, however, collect actual values for wetincome, including wetrents, dryrent,
gamingincomeandotherincome,whichwecanmakeuseof.However,thisdataisinsufficientto
produceabottomupcostingofthefreeoftiealternative.
It is worth noting at this point that several stakeholders interviewed by us have their doubts
whetheracomparablefreeoftieestimatecanexistconceptually.Onemajorpointdrawnfrom
theseconversationsisthatdryrentsarecalculatedbyRICSsurveyorsusingcomparablelocalpubs
as benchmarks, but the idea of comparable means that a tied pub would not be considered
comparable to a freeoftie pub and vice versa, because of the large number of (potentially
unquantifiable)countervailingeffects,suchaswetrents,SCORFA,andwhetherornotthetenant
wouldbecomeresponsibleformaintenanceinthefreeoftieoption.Wedoconsiderthatakey
risk to BIS in relation to this policy is exactly how such assessments will be carried out in a
transparentwaythatisnotoverlyreliantonassumptionsthatmaynotmaterialiseorchangeover
time.
The lack of a bottomup costing to estimate the impact of moving from a tied to a free of tied
state,orwhatleveloftransferwouldberequiredtodeliverthenoworseoffstatustargetedby
theproposedregulatorychange,meansthatwehavehadtoassumearangeoffeasiblevaluesfor
this transfer to test potential impacts. Providing an evidential base or a theoretical rationale for
theseisextremelychallenging,inpartbecause:
Identifyinganyfeasibledatasourcetousetocreateanestimatehasbeenchallenging
LondonEconomics
13
Conceptually, should we model a flat average transfer for all pubs, or should we scale
thisbysize?Simplyput,ifatenantinasmalltiedpubis5,000worseoffthanafreeof
tieequivalent,shouldhereceivelesstransferthanatenantsimilarlyworseoffinalarge
tiedpubthataccruesmorerevenue?Isitfairtoassumethattenantsinsmallerpubshave
asmallergapbetweentheircurrentpositionandanoworseoffpositionthanatenant
in a larger pub? It is actually easier to make the opposite argument, that in a large
revenuepub,thereismoreprofittoshare,sothetenantinalargetiedpubshouldhavea
smallergapbetweenhiscurrentincomeandanoworseoffincomethanalowrevenue
pub, where the pubco will still be looking to cover their costs, which may not be
correlatedwithrevenue:
Maintenancecostsareexpectedtobecorrelatedwithpubsquarefootage,which
may not be correlated with revenue, and depend on location and neighbouring
facilities. 29
Contract management costs may be a fixed cost, and therefore constant

irrespectiveofrevenue.
Thereisanargumentthatlargepubsalreadyaccruinglargerevenuesareableto
copewiththecoststhatfallonthemwithgreatereasethansmallerpubs,andit
may be logical therefore to argue that where pubcos provide support through
SCORFAtopubsthisislikelytobetargetedonsmallerpubs. 30
Debtservicingcostsarelikelytobecorrelatedtorevenue.
Ascanbeseen,thereisnounambiguousanswertothequestionwillthetransferfrompubcoto
tenantbecorrelatedwithrevenue?Wecannotevenaprioriidentifywhetherthisrelationship,if
itdoesexist,ispositiveornegative.
Candryrentsgonegative?Ifatenantinasmalltiedpubis5,000worseoffthanafree
oftie equivalent, but his dryrent is only 2,000 (possibly because in his case the
negotiated settlement is a high wetrent and low dryrent), then can dryrent only be
reducedtozero,orcanitgonegativetomakehimnoworseoff?
Ourapproach,therefore,isbasedoncomparingpubcorevenueandcostsinasimplebreakeven
analysis. At the margin, when we apply a transfer from the pubco to the tenant we can then
identify where pubs move from delivering positive net revenues to negative net revenues. This
analysis therefore identifies these pubs that are potentially likely to be divested under each
scenario.Thefollowingpointsarise:
In the baseline data there are already a number of pubs which have costs exceeding
revenuesandarethereforecandidatesfordivestment.Thisisbecausethepubsectoris
experiencing long term shifts in consumer behaviour that are reducing demand. We
excludethesefromourresultsasitwouldbeinaccuratetoallocateanyresultantclosures
fromwithinthisgrouptobeingcausedbythepolicy.Instead,weidentifyonlythosepubs
whichmovefrompositivenetprofitstonegativenetprofitsasaresultoftheintroduction
oftheabovedescribedpolicyoptions.
29
Twopubswiththesamefootprint,oneinacitycentreandtheotherinanisolatedvillage,willhaveverydifferentrevenues.
ThisargumentimpliesthegapbetweenwetrentandSCORFAinpubswithlowerrevenuesmaybelowerandwouldsupportthecase
thatweshouldscalebysize.
30
LondonEconomics
14
In the baseline data there are some pubs with negative income for the pubco. In
consultation with the pubcos we believe these to be those which are currently
transitioning from one tenancy to another, or otherwise in some form of special
measures. As this is a normal churn within the sector, including those which may be
divested,we havetakentheseintoaccountbyretaining theminoursample, aspartof
thebaseline.Astheyalreadydemonstratenegativeprofits,theywillnotshowachangein
stateasaresultofthepolicy.
3.3
Datacollection
3.3.1 Collectionprocess
LondonEconomicshasreceiveddatafromanumberofpubcos,withresponseseachcontaininga
randomsampleof300400anonymisedpubs.Thissamplehasbeendesignedasfollows:
We asked pubcos to provide a sample with 3040 pubs from each decile of their
populationofpubs,wherethepubsareorderedbyincomeforthepubco.
Weaskedpubcostoensurethattherewasageographicalspreadofpubsrepresentative
oftheirdistribution.
Informationcollectedfrompubcompaniesforeachpubwas:
Wetrent
Dryrent
Gamingrevenue
Otherrevenue
Rentpaidtofreeholderwherethepubcodoesnotowntheproperty
Managementandoperationcostsincurredbythepubcoinrelationtothepub
Maintenancecosts
Debtservicingcosts
Itisworthaddressingeachoftheseinmoredetail,asthedefinitionofeachisimportant:
Wetrentthisistakenastherevenueforthepubcofromwetsalesbythepub.Thiscan
benegativeasthisisanetfigure.Inthecasewherebeersalesarelow,forexampleifthe
pub is undergoing a period of maintenance, costs relating to the rental of the beer line
assetscanexceedtherevenues,whichweseeinasmallnumberofcasesforeachpubco.
Dryrentthisistakenasthepropertyrentalpaidbythetenanttothepubco.
Gaming revenue this is the income for the pubco from the gaming machines in the
pubs. The operational position on gaming income changed during the duration of our
study, with the pubcos voluntarily agreeing to remove the gaming income from the
divisiblebalancecalculationundertheirvoluntarycodes.Broadly,olderleaseswherethe
rent has not recently been reviewed may still be paying a rent that was assessed in a
different way, but income is now generally split three ways between the tenant, the
pubco and the machine supplier, and does not feed into divisible balance calculations.
However,ifthesiteisfreeoftieformachines,100%ofthemachineincomewasincluded
inthecalculationofrent,asnomachineincomewouldbetakenatsourcebythepubco.
OtherrevenueSpecialcasesandotherincomesforthepubcofromapub.
LondonEconomics
15
Rentpaidtofreeholderwherethepubcodoesnotownthepropertysomepubcos,as
well as using debt to finance the acquisition of pub properties, or indeed have reached
the point of having paid these debts off, also lease pub properties from a third party,
paying the rental on these in lieu of debt servicing costs. For pubcos that were able to
providethisinformationtousweallocatedtheleasecostsinplaceofdebtservicingcosts
for that pub. For those pubs that were not able to break this cost out, this cost was
includedinmanagementandoperationcostsandallocatedacrossallpubs.
ManagementandoperationcostsincurredbythepubcoinrelationtothepubThese
wereprovidedaseitheraggregatesforthewholefirm,orbrokendownonapubbypub
basis.
Maintenance costs These were provided as either aggregates for the whole firm, or
brokendownonapubbypubbasis.Thekeydifficultyisthatthecapitalcostsofthefirm
fallintothreemainparts:
Maintenanceonexistingpubs
Renovationofpubsintheestatetobringthembackintoaserviceablecondition
andremainapub
Purchasingnewpubs
Whilstthereislittleactivityinthethirdofthesecategories,thefirsttwoexistandthereis
anargumentthatweshouldattempttoexcludethecostsofrenovatingrundownpubs
fromthecostsincurredonserviceablepubsinoperation,asthiscostinnotreflectiveof
thetruecostsofoperatingandrunningthispubtoworkoutafairdivisionofprofits.We
haveincludedthiscostfortworeasons.First,inapracticalsense,itisnotpossiblefrom
the data we have to separate out this cost, and therefore we would need to apply a
roughandreadyassumptionofwhatshareofapubcocapitalcostsisrenovationcosts,
weakeningtherobustnessofouranalysis.Secondly,fromageneraltheoreticalpoint,we
cannot think of a pubs profitability and a fair allocation of profits without recognising
thatthereisatemporaldimensiontothisquestion,andthatalongtermpropertyowner
(thepubco)hastohavealongertermperspectivethanashorttermtenant.Thepubco
hastofundrenovationworkofthistypeorwatchadeclineintheirestate.Itcaneitherdo
this by laying aside cash every year into a sinking fund to pay these costs as they
emerge,orpaythesecostsastheyareincurred.Eitherway,thislongtermcosthastobe
fundedfromthepubcosrevenuestreamsfromitspubs.Thisappearstoustobethecore
differencebetweenthosewhoarguetenantshaveabaddealandthosewhoarguethe
pubcos are struggling; the first group conceptualise the pub as an operation in the
present, whereas the pubco conceptualises the pub as an operational unit of a larger
business, which also has to maintain an estate and fund corporate branding, marketing
andotherservices.Thissecondviewappearstoustobejustified,sowehaveallocatedall
capitalcoststothepubs.
Debt servicing costs These were provided as either aggregates for the whole firm, or
broken down on a pub by pub basis. Again here, in the same way as described with
maintenanceandrenovationcostsabove,thereisanargumenttosuggestthatthedebt
servicingcostsshouldbelimitedtointerestpayments,leavingtheprincipaltooneside.
This appears to us to suffer from the same arguments as presented above for the
maintenance costs. Whilst the pubco could only cover the interest charges from their
pubs, and indeed for those pubs that become unviable this may be a sensible firststep
approach, the longterm principal of the debt must be repaid for the pubco to reach a
position where it can significantly improve the longterm viability of the pubco by
significantlyreducingdebtservicingpayments.Indeed,duringthecurrentperiodoflow
LondonEconomics
16
and stable interest rates this is almost the optimal time to reduce debt levels before
interestratesrise.Inthecasewheredebtprincipalisnotpaidoff,ifinterestratesthen
rose we would expect to enter another cycle of divestments by pubcos, affecting the
longtermstabilityofthesector.Takingalongtermview,assuch,itisappropriateforus
toincludethefulldebtservicingcostsinthisanalysis.
Insomecases,negativeornullvaluesaregivenforrevenuefigures.Wehavebroughtthesetothe
attentionofpubcosandsoughtclarificationontheirinterpretation.Ingeneral,theseappeartobe
netcostsderivedbyapubneedingtopayservicechargeorbeerlinechargesforaperiodwhilsta
pub is not in operation (i.e. closed for maintenance or between tenants). In general, these
negativesaresmall.
3.3.2 Costallocation
Because many pubcos provided their cost data in aggregate form we carried out the following
stepstocreatepubleveldataforeachdataitem.
The aggregate costs are divided by the aggregate number of pubs owned by the firm, and
multipliedbackupbythenumberofpubsinthesampletoproduceasampleaggregatecost.
As a first step, management and operations costs, and maintenance costs are divided equally
between all pubs in the sample, except that the maintenance cost on leasehold pubs is down
weightedbyaquartertoreflectthetransferofsomemaintenancerisktothefreeholder.Thetotal
debt servicing cost is scaled according to the number of nonleasehold pubs in the sample and
dividedequallybetweenallnonleaseholdpubs.Forleaseholdpubs,weassumenodebtservicing
costs,however,wedoincludetheamountofrentpaidbythepubcotothefreeholderasacost.
Thisprocessdeliversanaggregateimputedcosttothepubcoforeachpubtobecomparedagainst
the aggregate declared revenue. This linear transformation of the cost data produces a larger
numberofunviablepubs,thanexpected.
3.3.3 Costcalibration
Expectedratesofunviability
UsingevidencefromtheBISconsultationdocumentthatnetpubclosuresare18aweek,froma
baseof51,000pubs,thissuggestsanannualnetclosurerateof939pubsayear,or1.8%ofthe
base.
As discussed above, assuming that tied pubs shut at the same rate as freehold and freeoftie
pubs,withtwothirdsofunviablepubsremainingclosed,thisimpliesanunviabilityrateof2.7%.
However, assuming pubcos take a longerterm view (demonstrated through renovation
expenditure etc), and with the capacity to support a pub for a short period of lower revenue
throughcoveringonlytheinterestonthedebtservicingcosts,wehavemadeanassumptionthata
LondonEconomics
17
pubco will only release a pub which is not able to return to profit within four years 31 . We thus
calculateabaseunviabilityrateof10.8% 32 .
Giventhatinaflatallocationofcoststopubs,farmorepubswouldbeunviablethanthistarget
number,wecalibratedthedatabyapplyinganadjustmentfactorontothecostallocation,scaled
accordingtotherevenueofeachpub,withthecostsmultipliedbytheadjustmentfactorforeach
pubsothathigherincomepubstakealargershareofthecosts.Thisisintuitive,ashigherincome
pubswouldbelikelytohavealargerfootfall,andbelocatedinhighcostareas(towncentresetc),
implying higher debt servicing costs (and maintenance costs). The key constraint on this re
allocation of costs to deliver the target unviability rate in the basecase is that the average
adjustmentfactormustbe1,sothatthetotalcostsacrossallpubsisequivalenttotheimputed
sampleaggregatecostbase
3.4
Modelsetup
3.4.1 Scenarios
Threeofthemodelinputshavebeenconstructedtotakearangeofvaluestoallowustogenerate
scenariostoreflecttheuncertaintyintheevidencewehavecollected.Thesethreefactorsareas
follows:
SCORFA
SCORFAtakesbasicvaluesof1,000,1,500,and2,000reflectingontheOFTestimateandanup
rateonthistoreflectinflationarypressuressince2010,andadownratetoaccountforanycost
savingefficienciesthepubcosmayhavebeenabletogenerateinthisperiod.
Inaddition,industrystakeholdersinterviewedforthepurposeofouranalysisindicatedarguments
infavourofanegativerelationshipbetweenthevalueofSCORFAapubcoprovidestoatenantand
the income which that tenants pub produces for the firm. Particularly, it appears that pubcos
grantlargerfinancialsupporttothosepubswhichaccruelessincomeandarethereforelesslikely
tobeabletocovertheexpensesincludedinSCORFAthemselves.Toaccountforthispossibility,
we divided the sampled pubs into three groups depending on their income (i.e. based on 33rd
incomepercentiles).Whereasthosewhoseincomerangedbetweenthe33rdand66thpercentileof
incomewouldsimplyreceiveoneofthebasicSCORFAvalueslistedabove,thoseinthelowincome
groupwouldreceive150%ofthisbasicvalue;incontrast,thoseinthehighincomegroupwould
beassignedonly50%ofthebasicvalue 33 .
Hence, to provide a realistic and wide range of SCORFA, we allow for variations in the basic
SCORFAestimatestoaccountforinflation,potentialpubcosavingefficiencies,aswellasindividual
pubincome.
31
Thefouryearassumptionisbasedonourconsultationwithstakeholders.
32
33
Toprovideanumericalexample,assumingabasicSCORFAvalueof1,000,alowincomepubwouldreceive1,500,amiddleincome
pub1,000,andapubwithhighincomeonly500inSCORFA.
LondonEconomics
18
Divisiblebalancepercentage
The divisible balance percentage is an estimate of the degree to which a pubco may be able to
increasethedryrentthatthetenantwouldfaceintheeventofthetenantgoingfreeoftieand
ceasingwetrentpayments.Thestandardapproachhereisthattheremovalofthewetrentwould
increase the divisible balance by an equivalent amount, and this assumption tests how large a
shareofthisdivisiblebalanceincreasethepubcowouldretainthroughitsdryrent,withvaluesof
35%,50%or65%.
So,asimpleworkedexamplewouldbe:
A tenant goes free of tie and ceases wetrent payments of 10,000 p.a. If the divisible
balance increase was 65%, this means the dryrent would increase by 6,500, and the
tenantwouldretain3,500.Weanticipatethepubcoswouldbeforcedtomaximisedry
rent increases in these circumstances not just on formerly tied tenants but also existing
freeoftietenantstobeabletosecuresufficientrevenuetocovercosts.Itisakeypoint
here that in this case the transfer of profits would not be from the pubco to the tied
tenant, but would be from the existing freeoftie tenant to the currently tied tenant,
although presently this is mitigated by the relevant pubcos having far fewer freeoftie
pubsthantheycurrentlyhavetiedpubs
Dryrentadjustment
Thedryrentadjustmentistheassumptionweusetocalculatethetransferfromthepubcotothe
tenant to deliver the no worse off condition. This cannot be estimated directly so instead we
have modelled four scenarios that represent the range of feasible possible values. Importantly,
thismeansthescenariospresentedinthisreportestimatethefullpossiblerangeofimpactsrather
than being direct estimates of what will happen. We tested using fixed values to estimate this
range(1,000,2,000,etc),butdeterminedthatpercentagesoftotalpubrevenuepresentedthe
most effective way of calculating this taking account of scale effects. We use values of 1%, 5%,
10%and15% 34 .
The divisible balance and dryrent adjustment in combination proxy for the potential outcomes
fromnegotiation,aswellastheuncertaintyaboutthepotentialofthepubtodelivernoworse
off. The scenarios use different assumptions about dry rent adjustments as a fraction of the
divisiblebalanceandthereforeattemptstoactastherangeofpotentialnegotiationoutcomes.As
such,weassumenegotiationwilloccurandwillbeseeninthescaleofthechangesagreed.Itis
importanttonotethatthereisaninbuiltinconsistencyaroundflexibilitytonegotiate.Theaimof
the policy is to deliver an outcome where the tenant is no worseoff than a freeoftie tenant.
Negotiation inherently delivers an outcome which is not the policy objective, but is some
compromisebetweenthecurrentallocationofprofitandanoworseoffoutcome.
34
This is not and does not attempt to disguise itself as a bottomup calculated figure. Our scenarios model the different potential
impactofanegotiationbetweenthetwoparties,workingfromatopdownassumptionthatifapubcolost15%ofitsrevenuefroma
pubitwouldstruggletokeepthatpubrunning.The15%isderivedfromthehighestaverageprofitmargin(profit/revenue)experienced
byanyofthepubcosinthedatasetsprovided;inshortitisanupperboundonwhatshareofthepubcosrevenuesthetenantcould
takebeforethenegotiationwouldstallasaresultofcompelledclosure.Wethereforeusethefourvaluestomerelyindicatetherange
ofoutcomeswhichoccurwithinthefeasibleset.
LondonEconomics
19
These three inputs can be combined to construct 36 distinct scenarios representing different
potentialoutcomes.
3.4.2 Modellingthesuggestedpolicyoptions
Wehavemodelledfivedistinctoptions
Statusquo
Revenue and costs remain the same and the percentage of unviable pubs is
calibratedto10.8%
Policyoption1withoutguestbeer
Dryrentisadjusteddownwardunderthenoworseoffrule
Policyoption1withguestbeeroption
Wetrentdecreasesby38%whichistheassumedguestbeershare 35
Dryrent increases by a proportion of the increase in the pubs divisible balance

(equalinthiscasetothedecreaseinwetrent)
Dryrent is adjusted downward under the no worse off rule using the dryrent
adjustmentassumption
Policyoption2whereallpubsgofreeoftie
WetrentiszeroforallpubsbutnowtenantshavetopaySCORFAandcapitalcosts
themselves 36
SCORFAisassumedtobe1.5timesmorecostlytotenantsthantopubcosdueto
economiesofscale
Dryrentincreasesusingthedivisiblebalanceassumption
Policyoption2wheresomepubsgofreeoftie
Pubsgofreeoftieifwhattheywouldgaininwetrentexceedswhattheywould
havetopayinSCORFAandcapitalcosts
Pubsthatgofreeoftiehavethesameoutcomeasinoption(d)andpubsthatdo
nothavethesameoutcomeasinoption1withoutguestbeer.
3.5
Employmenteffectsofclosure
Whenapubclosesarelativelysimpleequationcalculatesthenumberofjobslost.Usingdatafrom
theBBPAsBeerandPubSectorRegionalImpactAssessmentcompletedbyOxfordEconomics 37 ,
35
Itisassumedthatpubswillseektomaketheirtopsellingbrand(almostinvariablyalager)theirguestbeer.Whilstthepolicymaybe
designedtosupportcraftbeersandlocalrealalesweseethisasexceedinglyunlikely.Datawehaveseenfromexpertsourcessuggest
standardlagersalesarenormallybetween35%and40%ofwetsales
36
Taking an example of a pubco that retains responsibility for repairs to the pub, and which delivers this through a large facilities
managementoperationtoseveralthousandpubs,wecanseethevaluetothetenantofthisservice,butwecannoteasilyimputethe
costforafreeoftietenant(whomayultimatelybeaskedtotakethisresponsibilityonthemselvesandwouldbeforcedtopayinthe
openmarket).Assuch,weconsiderthatfortheoptionswherethetiedpubremainsatiedpub(theactivationofthecode),therewould
benochangeinthecontractualmodelandthepubcowouldstillincurthemaintenancecosts.Forthefreeoftieoption,wewouldneed
toattempttomodeltheincreaseincostsapubwouldfaceifitwasattemptingtopurchasetheseservicesthemselves,asunderthe
standardfreeoftiecontractualterms.Thiswouldhaveakeysecondaryeffectinthatweshouldassumethatcostsavingsdeliveredby
thepubcosbuyingpowertoanytiedpubswhichremaintiedpubswouldfall,butwouldfalllessthanproportionatelyduetothepubco
losingeconomiesofscaleandthereforebuyingpower,whichwouldimplyhigherunitcosts.Othersectorswehavereviewedsuggest
thatadropindemandforaserviceof7%wouldincreaseunitpricesbyaround1%,butwehavenotincludedthisinouranalysistoas
wefeltitwasimportanttofocusonthekeysecondaryeffectofdemandshifting.
LondonEconomics
20
which identified an average of five fulltime direct employees per pub 38 . This figure excludes
indirect employment, but given many of these are in support industries that may not reduce in
scale (a delivery van still needs the same driver and crewman even if they undertake one less
delivery), we have decided to not take into account the potential indirect labour effects, or the
secondordereffectsofthealternativeuseofthepropertyorthealternativespendingofthe40%
ofsalesthatmovesoutsidethepubssector.
37
OxfordEconomics,2012
38
NotethattheoriginaldataprovidefulltimedirectemploymentfortheUKpubandbreweryindustry,intotalandbyregion.Forthe
purposeofouranalysis,basedonthegeographicalrangeofthesuggestedpolicyoptionswetookasimpleaverageoffulltimedirect
employmentperpubacrossallEnglishregionsandWales,thusexcludingScotlandandNorthernIreland.
LondonEconomics
21
4Resultsandconclusions
Resultsandconclusions
4.1
Whymightapubclose?
Theproposalsunderdiscussionaregenerallycharacterisedassimplyreallocatingprofitfromthe
pubcompany(pubco)tothetenant,soifthereisonlyaredirectionofprofitfromoneagentto
another,whyshouldthischangetheviabilityofthepub?Theansweristhatthediscussiondoes
notrevolvearoundprofitbutratheraroundoperatingmargin,whichisadifferentconcept.
Profit is the residual from revenue after all costs have been covered. Operating margin is the
revenueremainingafterfrontlineoperationalcostshavebeenpaid.Thedifferenceisthatwhilsta
pub may be operationally viable in terms of the revenue exceeding the operating costs and the
costofsalegoodsthatarefacedbythetenant,therearefurthercoststhatmustbeconsidered,
which are currently borne by the pubco, including, as explained above, capital costs (e.g.
maintenance), and debt servicing costs (i.e. the interest on the debt used to purchase the
property).
coverthesecosts,andthemanagementcostswhichthepubcofacesaroundtheseelementsand
thecontractwiththetenant,aswellascoveringthecostofanySCORFAprovisions.Iftheincome
from the pub falls below the aggregate of these, the pub may continue to appear viable to the
tenantbutwouldnolongerbeviableforthepubco,whichwouldbeforcedtodivestthisasset.
4.2
Wehavemodelledfiveoptions:
Thecurrentbasecase
ImplementingtheCode,butnotallowingaguestbeeroption
ImplementingtheCodewiththeguestbeeroption
Allowingtenantstogofreeoftie,withalltenantstakingthisoption
Allowing tenants to go freeoftie, but only doing so if it improved their financial
position 39 .
Wehavemodelledtheseusing36scenarios,basedonthekeyvariablesofthevalueofSCORFA,
the divisible balance percentage and the dryrent adjustment. The following tables present the
numberofpubswhichbecomeunviable,thenumberofpubsweexpecttocloseandthenumber
ofjobsweexpecttobelostasaresultofthepolicyandperscenario,takingintoaccountonlythe
firstordereffects.
4.2.1 Firstordereffects
39
ThevastmajorityoffreeoftiepubsareresponsibleformaintenanceanddonotreceiveSCORFA,whichmeanstheadditionalincome
frombeingfreeofthewetrentneedstoexceedthesumoftheseforatenanttowishtousetheoption.
LondonEconomics
22
Table4: Numberoftiedpubsbecomingunviableanddivestedbypubcos(firstordereffects)
underdifferentscenarios
Scenario
SCORFA
value()
Divisible
balance
value
Dryrent
adjustment
value
Option1
noguest
beer
Option1
guestbeer
Option2
freeoftie
(all)
Option2
freeoftie
(optimised)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
1,000
1,000
1,000
1,000
1,000
1,000
1,000
1,000
1,000
1,000
1,000
1,000
1,500
1,500
1,500
1,500
1,500
1,500
1,500
1,500
1,500
1,500
1,500
1,500
2,000
2,000
2,000
2,000
2,000
2,000
2,000
2,000
2,000
2,000
2,000
2,000
35%
35%
35%
35%
50%
50%
50%
50%
65%
65%
65%
65%
35%
35%
35%
35%
50%
50%
50%
50%
65%
65%
65%
65%
35%
35%
35%
35%
50%
50%
50%
50%
65%
65%
65%
65%
1%
5%
10%
15%
1%
5%
10%
15%
1%
5%
10%
15%
1%
5%
10%
15%
1%
5%
10%
15%
1%
5%
10%
15%
1%
5%
10%
15%
1%
5%
10%
15%
1%
5%
10%
15%
2,300
4,100
6,800
7,200
2,300
4,100
6,800
7,200
2,300
4,100
6,800
7,200
2,300
4,100
6,800
7,200
2,300
4,100
6,800
7,200
2,300
4,100
6,800
7,200
2,300
4,100
6,800
7,200
2,300
4,100
6,800
7,200
2,300
4,100
6,800
7,200
8,000
9,000
10,400
11,500
6,800
8,200
9,500
10,700
5,700
7,300
8,700
9,900
8,000
9,000
10,400
11,500
6,800
8,200
9,500
10,700
5,700
7,300
8,700
9,900
8,000
9,000
10,400
11,500
6,800
8,200
9,500
10,700
5,700
7,300
8,700
9,900
9,500
9,500
9,500
9,500
8,500
8,500
8,500
8,500
6,900
6,900
6,900
6,900
9,400
9,400
9,400
9,400
8,400
8,400
8,400
8,400
6,900
6,900
6,900
6,900
9,200
9,200
9,200
9,200
8,300
8,300
8,300
8,300
6,900
6,900
6,900
6,900
9,600
9,700
9,900
10,000
8,600
8,700
8,900
9,000
7,000
7,100
7,300
7,300
9,500
9,600
9,800
9,900
8,500
8,600
8,800
8,900
7,000
7,100
7,300
7,400
9,400
9,500
9,800
9,900
8,400
8,500
8,800
8,900
6,900
7,100
7,300
7,400
Source:LondonEconomics'analysis
Thenexttabletakesaccountofthefactthatcurrentbestestimatessuggestthataroundtwothirds
ofthosepubsdivestedbypubcoswillclose,withonethirdbeingreopenedasapubbyadifferent
agent. The table above, therefore, shows the impact on the pubcos, and the table below shows
theimpactonthepubsectorasawhole.
LondonEconomics
23
Table5: TotalnumberoftiedPubsclosingasaresultofthepolicy(firstordereffects)
Scenario
SCORFA
value()
Divisible
balance
value
Dryrent
adjustment
value
Option1
noguest
beer
Option1
guestbeer
Option2
freeoftie
(all)
Option2
freeoftie
(optimised)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
1,000
1,000
1,000
1,000
1,000
1,000
1,000
1,000
1,000
1,000
1,000
1,000
1,500
1,500
1,500
1,500
1,500
1,500
1,500
1,500
1,500
1,500
1,500
1,500
2,000
2,000
2,000
2,000
2,000
2,000
2,000
2,000
2,000
2,000
2,000
2,000
35%
35%
35%
35%
50%
50%
50%
50%
65%
65%
65%
65%
35%
35%
35%
35%
50%
50%
50%
50%
65%
65%
65%
65%
35%
35%
35%
35%
50%
50%
50%
50%
65%
65%
65%
65%
1%
5%
10%
15%
1%
5%
10%
15%
1%
5%
10%
15%
1%
5%
10%
15%
1%
5%
10%
15%
1%
5%
10%
15%
1%
5%
10%
15%
1%
5%
10%
15%
1%
5%
10%
15%
1,500
2,700
4,600
4,800
1,500
2,700
4,600
4,800
1,500
2,700
4,600
4,800
1,500
2,700
4,600
4,800
1,500
2,700
4,600
4,800
1,500
2,700
4,600
4,800
1,500
2,700
4,600
4,800
1,500
2,700
4,600
4,800
1,500
2,700
4,600
4,800
5,300
6,000
6,900
7,700
4,600
5,500
6,400
7,100
3,800
4,900
5,800
6,600
5,300
6,000
6,900
7,700
4,600
5,500
6,400
7,100
3,800
4,900
5,800
6,600
5,300
6,000
6,900
7,700
4,600
5,500
6,400
7,100
3,800
4,900
5,800
6,600
6,400
6,400
6,400
6,400
5,700
5,700
5,700
5,700
4,600
4,600
4,600
4,600
6,200
6,200
6,200
6,200
5,600
5,600
5,600
5,600
4,600
4,600
4,600
4,600
6,100
6,100
6,100
6,100
5,500
5,500
5,500
5,500
4,600
4,600
4,600
4,600
6,400
6,500
6,600
6,700
5,700
5,800
5,900
6,000
4,700
4,700
4,800
4,900
6,300
6,400
6,600
6,600
5,700
5,700
5,900
5,900
4,600
4,700
4,900
4,900
6,200
6,300
6,500
6,600
5,600
5,700
5,800
5,900
4,600
4,700
4,900
5,000
LondonEconomics
24
Table6: Joblossesfrompubclosures(firstordereffects)
Scenario
SCORFA
value()
Divisible
balance
value
Dryrent
adjustment
value
Option1
noguest
beer
Option1
guestbeer
Option2
freeoftie
(all)
Option2
freeoftie
(optimised)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
1,000
1,000
1,000
1,000
1,000
1,000
1,000
1,000
1,000
1,000
1,000
1,000
1,500
1,500
1,500
1,500
1,500
1,500
1,500
1,500
1,500
1,500
1,500
1,500
2,000
2,000
2,000
2,000
2,000
2,000
2,000
2,000
2,000
2,000
2,000
2,000
35%
35%
35%
35%
50%
50%
50%
50%
65%
65%
65%
65%
35%
35%
35%
35%
50%
50%
50%
50%
65%
65%
65%
65%
35%
35%
35%
35%
50%
50%
50%
50%
65%
65%
65%
65%
1%
5%
10%
15%
1%
5%
10%
15%
1%
5%
10%
15%
1%
5%
10%
15%
1%
5%
10%
15%
1%
5%
10%
15%
1%
5%
10%
15%
1%
5%
10%
15%
1%
5%
10%
15%
7,500
13,500
22,600
23,700
7,500
13,500
22,600
23,700
7,500
13,500
22,600
23,700
7,500
13,500
22,600
23,700
7,500
13,500
22,600
23,700
7,500
13,500
22,600
23,700
7,500
13,500
22,600
23,700
7,500
13,500
22,600
23,700
7,500
13,500
22,600
23,700
26,500
29,900
34,300
38,100
22,700
27,200
31,600
35,500
18,800
24,300
28,900
32,800
26,500
29,900
34,300
38,100
22,700
27,200
31,600
35,500
18,800
24,300
28,900
32,800
26,500
29,900
34,300
38,100
22,700
27,200
31,600
35,500
18,800
24,300
28,900
32,800
31,600
31,600
31,600
31,600
28,200
28,200
28,200
28,200
23,000
23,000
23,000
23,000
31,100
31,100
31,100
31,100
27,900
27,900
27,900
27,900
23,000
23,000
23,000
23,000
30,600
30,600
30,600
30,600
27,500
27,500
27,500
27,500
23,000
23,000
23,000
23,000
31,900
32,300
32,900
33,100
28,500
28,800
29,400
29,700
23,200
23,500
24,100
24,300
31,500
31,800
32,600
32,800
28,200
28,500
29,300
29,500
23,100
23,500
24,300
24,500
31,000
31,600
32,400
32,700
27,700
28,200
29,100
29,400
23,000
23,500
24,400
24,700
Table 7 shows a summary of the range of the share of the current stock of 51,000 pubs we
estimate would close under each option, over and above those which would close in the base
case. We have modelled a number of scenarios to reflect the diversity of opinion amongst
stakeholdersinrelationtocertainkeyassumptions.
LondonEconomics
25
Table7: Theimpactonpubclosuresandemployment(firstordereffectsonly)
Option1no
guestbeer
Option1
guestbeer
Option2free
oftie(all)
Option2free
oftie
(optimised)
514%
1123%
1419%
1420%
1136%
2858%
3548%
3550%
1,5004,800
7,50023,700
3,8007,700
18,80038,100
4,6006,400
23,00031,600
4,6006,700
23,00033,100
Minmaxranges
Percentageofallpubs
closing
Percentageoftiedpubsin
scopeofCodeclosing
Numberofpubsclosing
Numberofjobslost
However,Table7onlyaccountsforthefirstordereffects,ordirecteffectsofthechange.With
this number of pubs closing we need to also consider the second order effects, or knockon
implications of the change. The most important of these is the behaviour of consumers who
previouslywouldhaveusedaclosedpubandwhonowhaveachoice:
Stopdrinking
Stopusingapubanddrinkatadifferentvenue(home,club,baretc)
Movetoadifferentpub
The literature on pubswitching behaviour is very thin. We have assumed that only 60% of
customerswhosepubshutmovetoanotherpub.Wehaveaddedthisdemand(andthereforewet
revenues)tothepubsmodelledtoseewhatimpactwecouldexpectthistohave.Asshownbelow
thisactstoreducetherangeofpubsestimatedasclosing.
4.2.2 Whydoestheguestbeeroptiongivetheworstclosureestimates?
Inrelationtowhytheguestbeeroptionappearstogivetheworstresults,intermsofpubclosures,
thisisbecausetheintroductionofaguestbeeroptionleadstodistinctdifferencesinseveralofthe
keyparameterswhichaffecttheprofitwhichapubcocanderivefromagiventenantunderPolicy
OptionI(IntroductionofaStatutoryCodeandIndependentAdjudicator):
Undertheguestbeeroption,comparedtothescenariowithoutthisoption,itisassumedthatthe
tenantwillbeallowedtobuyitsstandardlager(whichconstitutesanestimated38%oftotalbeer
sales)fromasupplierotherthanthepubcompanyitself.Asaresultofthisswitchingofsomebeer
purchases to another supplier, the wet rent which the pub company receives from the given
tenantreducesby38%,i.e.thewetrentreceivedbythepubcoislowerundertheguestbeeroption
thanotherwise.
Inordertopartiallycompensateforthislossinwetrent,itisassumedthepubcowillincreasethe
dry rent (before adjustments under the Code) which it charges to the tenant. In particular, it is
assumedthatthepubcompanyincreasesitsdryrentbyagivenpercentage(35,50or65%)ofthe
differenceinthewetrentpaidbythetenantwithoutandwiththeguestbeeroption.Particularly,
since the tenant pays a lower wet rent to the pub company under the guest beer option, this
increasesthedivisiblebalanceaccruedbythetenant.InaccordancewiththeBISapproachtaken
intheconsultationdocument,thepubcotakesashareofthisincreaseinthedivisiblebalancein
LondonEconomics
26
dryrent,addingittothedryrentwhichthetenantwouldhavetopayunderthenoguestbeer
option.
Summarising these effects, whereas the pubco receives a lower amount of wet rent from the
tenantundertheguestbeeroption,itpartiallyrecoversthislossinwetrentbychargingahigher
dryrent(beforetheadjustmentrequiredunderthenewCode)thanunderthesituationwithno
guest beer option. As a result, the total revenue accrued by the pubco from a given pub before
takingaccountofthenecessarydryrentadjustmentrequiredundertheCodeislowerunderthe
guestbeerscenario.
As a result, for every pub and every scenario considered in our model, the revenue which the
pubco accrues from a given pub is lower when introducing a guest beer option along with the
StatutoryCodeandAdjudicator.Sincethemodelledcoststothepubcoarethesameunderboth
scenarios,thisimpliesthattheprofitstothepubcofromagivenpubarealsolowerforeverypub
andscenariowhenconsideringtheguestbeeroption.Hence,whereasagivenpubmightcurrently
beatthemarginofviabilityforthepubcompany,theintroductionoftheguestbeeroptionmight
resultinanegativeprofitforthepubcoforthesamepub,thusmakingitunviableforthepubco.
4.2.3 Secondordereffects
Thefollowingtablesoutlinethemodellednumberofpubsbecomeunviable,thoseclosingandthe
resultantjoblossesoncesecondordereffectsaretakenintoaccount.
Wetestedscenariosusingbetween60%and90%retentionofdrinkersinthepubsectorandthe
resultsonlychangedmarginally.Thisdemonstrateshowmanypubs,giventhedatawecanaccess
and the assumption we have made, are close to the margin of viability, such that the changes
modelledinthefirstroundeffectsarelargeenoughtocausesignificantnumbersofclosures,but
thatalsothesecondordereffectofafewpercentincreaseindemandissufficienttobringmanyof
them back into viability again. This is a key result; how potentially volatile the number of pubs
closingistoassumptionchanges.
LondonEconomics
27
Table8: Tiedpubsbecomingunviableanddivestedbypubcos(secondordereffects)
Scenario
SCORFA
value()
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
1,000
1,000
1,000
1,000
1,000
1,000
1,000
1,000
1,000
1,000
1,000
1,000
1,500
1,500
1,500
1,500
1,500
1,500
1,500
1,500
1,500
1,500
1,500
1,500
2,000
2,000
2,000
2,000
2,000
2,000
2,000
2,000
2,000
2,000
2,000
2,000
Divisible
balance
value
35%
35%
35%
35%
50%
50%
50%
50%
65%
65%
65%
65%
35%
35%
35%
35%
50%
50%
50%
50%
65%
65%
65%
65%
35%
35%
35%
35%
50%
50%
50%
50%
65%
65%
65%
65%
Dryrent
adjustment
value
1%
5%
10%
15%
1%
5%
10%
15%
1%
5%
10%
15%
1%
5%
10%
15%
1%
5%
10%
15%
1%
5%
10%
15%
1%
5%
10%
15%
1%
5%
10%
15%
1%
5%
10%
15%
Option 1 Option 1 Option 2 Option 2

no guest guestbeer
freeoftie
freeoftie
beer
(all)
(optimised)
100
800
1,000
100
800
1,000
100
800
1,000
100
800
1,000
100
800
1,000
100
800
1,000
100
800
1,000
100
800
1,000
100
800
1,000
1,400
1,800
2,200
2,400
1,100
1,600
1,900
2,200
600
1,200
1,700
2,000
1,400
1,800
2,200
2,400
1,100
1,600
1,900
2,200
600
1,200
1,700
2,000
1,400
1,800
2,200
2,400
1,100
1,600
1,900
2,200
600
1,200
1,700
2,000
2,000
2,000
2,000
2,000
1,500
1,500
1,500
1,500
1,100
1,100
1,100
1,100
1,900
1,900
1,900
1,900
1,500
1,500
1,500
1,500
1,100
1,100
1,100
1,100
1,800
1,800
1,800
1,800
1,500
1,500
1,500
1,500
1,100
1,100
1,100
1,100
2,000
2,000
2,100
2,100
1,600
1,600
1,600
1,700
1,100
1,200
1,200
1,200
2,000
2,000
2,100
2,100
1,500
1,500
1,600
1,600
1,100
1,200
1,200
1,300
1,900
1,900
2,000
2,000
1,500
1,500
1,600
1,700
1,100
1,100
1,200
1,300
LondonEconomics
28
Table9: Totalnumberoftiedpubsclosing(secondordereffects)
Scenario
SCORFA
value()
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
1,000
1,000
1,000
1,000
1,000
1,000
1,000
1,000
1,000
1,000
1,000
1,000
1,500
1,500
1,500
1,500
1,500
1,500
1,500
1,500
1,500
1,500
1,500
1,500
2,000
2,000
2,000
2,000
2,000
2,000
2,000
2,000
2,000
2,000
2,000
2,000
Divisible
balance
value
35%
35%
35%
35%
50%
50%
50%
50%
65%
65%
65%
65%
35%
35%
35%
35%
50%
50%
50%
50%
65%
65%
65%
65%
35%
35%
35%
35%
50%
50%
50%
50%
65%
65%
65%
65%
Dryrent
adjustment
value
1%
5%
10%
15%
1%
5%
10%
15%
1%
5%
10%
15%
1%
5%
10%
15%
1%
5%
10%
15%
1%
5%
10%
15%
1%
5%
10%
15%
1%
5%
10%
15%
1%
5%
10%
15%

no guest guestbeer
freeoftie
freeoftie
beer
(all)
(optimised)
100
500
700
100
500
700
100
500
700
100
500
700
100
500
700
100
500
700
100
500
700
100
500
700
100
500
700
1,000
1,200
1,400
1,600
700
1,000
1,300
1,500
400
800
1,100
1,300
1,000
1,200
1,400
1,600
700
1,000
1,300
1,500
400
800
1,100
1,300
1,000
1,200
1,400
1,600
700
1,000
1,300
1,500
400
800
1,100
1,300
1,300
1,300
1,300
1,300
1,000
1,000
1,000
1,000
700
700
700
700
1,300
1,300
1,300
1,300
1,000
1,000
1,000
1,000
700
700
700
700
1,200
1,200
1,200
1,200
1,000
1,000
1,000
1,000
700
700
700
700
1,300
1,400
1,400
1,400
1,000
1,100
1,100
1,100
800
800
800
800
1,300
1,300
1,400
1,400
1,000
1,000
1,100
1,100
800
800
800
800
1,200
1,300
1,300
1,400
1,000
1,000
1,100
1,100
700
800
800
800
LondonEconomics
29
Table10:Joblossesasaresultofpubclosures(secondordereffects)
Scenario
SCORFA
value()
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
1,000
1,000
1,000
1,000
1,000
1,000
1,000
1,000
1,000
1,000
1,000
1,000
1,500
1,500
1,500
1,500
1,500
1,500
1,500
1,500
1,500
1,500
1,500
1,500
2,000
2,000
2,000
2,000
2,000
2,000
2,000
2,000
2,000
2,000
2,000
2,000
Divisible
balance
value
35%
35%
35%
35%
50%
50%
50%
50%
65%
65%
65%
65%
35%
35%
35%
35%
50%
50%
50%
50%
65%
65%
65%
65%
35%
35%
35%
35%
50%
50%
50%
50%
65%
65%
65%
65%
Dryrent
adjustment
value
1%
5%
10%
15%
1%
5%
10%
15%
1%
5%
10%
15%
1%
5%
10%
15%
1%
5%
10%
15%
1%
5%
10%
15%
1%
5%
10%
15%
1%
5%
10%
15%
1%
5%
10%
15%

no guest guestbeer
freeoftie
freeoftie
beer
(all)
(optimised)
500
2,700
3,500
500
2,700
3,500
500
2,700
3,500
500
2,700
3,500
500
2,700
3,500
500
2,700
3,500
500
2,700
3,500
500
2,700
3,500
500
2,700
3,500
4,700
6,000
7,200
8,100
3,600
5,100
6,400
7,400
2,100
4,100
5,700
6,600
4,700
6,000
7,200
8,100
3,600
5,100
6,400
7,400
2,100
4,100
5,700
6,600
4,700
6,000
7,200
8,100
3,600
5,100
6,400
7,400
2,100
4,100
5,700
6,600
6,500
6,500
6,500
6,500
5,100
5,100
5,100
5,100
3,700
3,700
3,700
3,700
6,400
6,400
6,400
6,400
5,000
5,000
5,000
5,000
3,700
3,700
3,700
3,700
6,100
6,100
6,100
6,100
4,900
4,900
4,900
4,900
3,700
3,700
3,700
3,700
6,600
6,700
6,900
7,000
5,200
5,300
5,400
5,500
3,800
3,800
4,000
4,100
6,500
6,600
6,900
7,000
5,000
5,100
5,400
5,500
3,700
3,800
4,100
4,200
6,200
6,400
6,600
6,800
4,900
5,000
5,300
5,500
3,700
3,800
4,100
4,200
LondonEconomics
30
Table11: Theimpactonpubclosuresandemploy(secondroundeffects)
Minmaxranges
Percentageofall
pubsclosing
Percentageoftied
pubsinscopeof
codeclosing
Numberofpubs
closing
Numberofjobslost
Option1no
guestbeer
Option1guest
beer
Option2freeof
tie(all)
01.4%
0.93.3%
05.3%
3.212.3%
5.79.9%
5.510.7%
0700
4001,600
8001,300
7001,400
03,500
2,1008,100
3,7006,500
3,7007,000
1.52.6%
Option2freeof
tie(optimised)
1.52.8%
4.2.4 Whydoesthenoguestbeeroptiongivesuchlownumbersofclosures?
Wherethetablesindicateavalueofzero,thisisbecausethenumberofclosureswaslowerthanin
thebaseline(i.e.negativeclosuresasaresultofthepolicy).Thisiscausedbythefollowing:
Inthebaseline,toreflectrecent changestakenforwardbythepubcos,weremovedthe
gamingincomefromthedivisiblebalancecalculation.
Unfortunately, there is a noticeable number of cases where the pubcos gaming income
was negative (costs exceeded revenue, as gaming income is a net figure). In these cases
thepubcossacrificingofgamingincomemakesthetenantworseoff.
Insuchcases,mostclearlyexposedinoption1whensmalltransfersaremadetothetenantunder
thenoworseoffpolicy(i.e.thepolicyimpactisthesmallest),theneteffectofthetwochangesis
toleavetenantincomelowerthanpresentlyforpubswithnegativegamingincomeforthepubco.
Theimplicationofthisisthatinsuchcasespubcoincomesincreaseundertheintroductionofthe
codeandpubsbecomemoreviable,thusreducingclosuresbytransferringoperatingmarginfrom
tenants to pubcos the exact opposite of the policy intent. Because we anticipated that this
would be prevented from actually happening through negotiation we set cases where total
closureswentnegativetozeroclosures,ontheassumptions
Finally, the second secondorder effect, which could occur, is that pubcos could reduce their
managementcostsandmaintenancebudgetsandrefinancetheirdebttoreducetheircosts.We
anticipate they would attempt to do this, but given the degree to which the pubcos have faced
lossesinrecentyears,webelievetheymaynowfinditdifficulttogosignificantlydeeperinterms
of efficiencies. As such, we have not modelled this effect. In part, we have not attempted to
estimatethiseffectbecausebothclosureandpubsturningfreeoftiecouldreducetheeconomies
of scale that pubcos currently achieve, driving in additional cost pressures on those tied pubs
whichremain.
LondonEconomics
31
4.3
Conclusions
ThetiedpubmodelisoneofthemostcomplexindustrialrelationshipsremaininginUKindustry
and aims to deliver a system to ensure floor levels of demand for British brewers, sustaining
diversityandthetraditionalfamilybrewerymodel.Inrecenttimes,thishasbecomeincreasingly
necessary, as the consumption of beer in pubs has declined, with the numbers of barrels sold
falling dramatically as consumption patterns have changed, both due to lower prices in
supermarketsandtheemergenceofalternativessuchasrestaurants,clubs,andbars.Anumberof
stakeholders interviewed noted that the UK is probably still operating excess pub supply of
approximately6,000pubs,suggestingasustainablenumberofpubsofapproximately45,000.
Thisoversupplyhasled tolowprofitability,bothformanytenantsandpubcos,particularlyina
climatewhereservicingdebthasbecomedifficult.Thekeyfindingofthisstudyisnotthenumber
ofpubswhichmaycloseasaresultofonepolicyoranother,butratherthehighnumberofpubs
that currently appear to be at the margin of viability. Irrespective of what changes may be
proposed or considered, the interlocking nature of a large variety of revenuestreams, and the
high level of costs being faced by pubcos, suggest that almost any policy reform may have
noticeableandunpredictableeffects.
Intheestimatesweproduceoftheimpactoftheconsultationproposals,eventakingaccountof
secondordereffects,ofthe13,300pubswebelievewillbeinscopeoftheCode,upto2,400or18
percent could become unviable for their pubco owners, on top of those already unviable
(c.1,300) 40 within the base case scenario, although we estimate a third of these would reopen
underalternativemanagement.
The threat that the number of tied pubs currently operated by the pubcos may diminish by
between 2530% due to closures, even if these pubs subsequently reopen under a different
owner,anddisregardinganywhomaywishtotakeupthefreeoftieoption,maybesufficientto
eliminatetheeconomiesofscaleinpurchasingthatmanytiedtenantsinthissector(unknowingly)
benefitfrom.Thissuggeststhatthereisarealpossibilitythateachoftheproposedpolicyreforms,
exceptpossiblythecodewithoutpermittingguestbeer,insteadofdeliveringthepolicyobjective
ofensuringtiedtenantsaretreatedfairly,i.e,noworseoffthanfreeoftietenants,mayleadto
theendofalargescaletiedpubsystem.
This,however,maynotbeasdisastrousasitinitiallysounds.Thetiedpubmodelisverysimilarto
amorestandardfranchisingarrangement,albeitonewherethepubcosuppliesthesalesproduct
andleasesthepropertytothefranchisee.Thismodelisalreadybeingattemptedbysomepubcos
andmaybeworthyoffurtherconsideration.
Intheshortterm,itisalsoworthnotingthatthePICAS/PIRRSsystem,whichisthemostconcrete
element of the current voluntary framework, is now becoming an established part of the
infrastructure, and appears to be having some effect in starting to address the worst tenant
circumstances 41 ,althoughsomestakeholdershaveraisedthepointwithusthatsometenantsare
wary of taking the PICAS/PIRRS route in case the discretion available ends up increasing their
rental.
40
Itmustbenotedthatthisfigureisdrivenbyourassumptionsinthecalibrationexerciseandisthereforeentirelyanartefactofour
modellingprocess.
41
Business,InnovationandSkillsCommittee(2011).
LondonEconomics
32
It is our conclusion that the reforms proposed in the consultation will close up to 1,600 pubs,
although there is very great uncertainty about the precise value; In particular, the size of the
transfer from pubcos to tenants resulting from the no worse off principle is very hard to
estimate, our results reflect the impact of a range of possible transfer values; However, there is
clearlysurpluspubcapacity,inquiteavolatilemarketwhere,withsomanypubsonthemarginof
viabilityitishardtodeterminewhichpubswillclose.Stakeholdersraisedwithusthatthatthere
maybeupto6,000surpluspubsintheUKforming12%ofthemarket,andifthispolicydiddeliver
severalthousandclosedpubsitwouldactasasubstantialfractionofthislongtermtrendwhichis
likelytooccurunlessmajorchangestotaxpolicyandsocialnormstakeplace.
If one assumes, as we have done, 60% of consumers move toanother pub that implies that, on
averagepubswhichremainwillseefootfalls7.2%higherthanpresent.Thiswouldbesufficientto
turn a poorly performing pub into a more attractive prospect if it can see the immediate future
out.Assuchitmaydeliverenoughofaboosttootherpubstoreduceclosureratesinthemedium
term.
LondonEconomics
33
5References
References
BusinessandEnterpriseCommittee(2009).Pubcompanies,seventhreportofSession200809.
http://www.publications.parliament.uk/pa/cm200809/cmselect/cmberr/26/26i.pdf[Accessed3
May].
Business,InnovationandSkillsCommittee(2010).Pubcompanies,followup:Government
responsetotheCommitteesfifthreportofSession200910.
http://www.publications.parliament.uk/pa/cm200910/cmselect/cmbis/503/503.pdf[Accessed3
May2013].
Business,InnovationandSkillsCommittee(2011).Pubcompanies,tenthreportofSession2010
12.
http://www.publications.parliament.uk/pa/cm201012/cmselect/cmbis/1369/136902.htm
[Accessed25June2013].
DepartmentforBusiness,InnovationandSkills(2012).CallforevidenceletterfromVinveCableto
RodgerVickers:Selfregulationoftheindustryframeworkandcompanycodes.
https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/70009/self
regulationoftheindustryframeworkandcompanycodes.pdf[Accessed6May2013].
DepartmentforBusiness,InnovationandSkills(2013a).Pubcompaniesandtenants:a
governmentconsultation.Availableat:
https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/193711/13718
pubcompaniesandtenantsconsultation.pdf[Accessed3May2013].
DepartmentforBusiness,InnovationandSkills(2013b).Pubcompaniesandtenants:impact
assessment.Availableat:
https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/192046/13738
pubcompaniesandtenantsimpactassessment.pdf[Accessed3May2013].
OfficeofFairTrading(2010).CAMRAsupercomplaintOFTfinaldecision.
http://www.oft.gov.uk/shared_oft/supercomplaints/OFT1279.pdf[Accessed3May2013].
OxfordEconomics(2012).BeerandpubsectorUKregionalimpactstudy.
http://www.beerandpubjobs.co.uk[Accessed25June2013].
RoyalInstitutionofCharteredSurveyors(2010).Thecapitalandrentalvaluationofpublichouses,
bars,restaurantsandnightclubsinEnglandandWales.Availableat:
http://www.bii.org/documents/703[Accessed25June2013].
TradeandIndustryCommittee(2004).Pubcompanies,secondreportofSession200405.
http://www.publications.parliament.uk/pa/cm200405/cmselect/cmtrdind/128/128i.pdf[Accessed
3May2013].
LondonEconomics
34

Tied Pubs Final Report

Загружено:

Сведения о документе

Авторское право

Доступные форматы

Поделиться этим документом

Поделиться или встроить документ

Параметры публикации

Этот документ был вам полезен?

Это неприемлемый материал?

Авторское право:

Доступные форматы

Tied Pubs Final Report

Загружено:

Авторское право:

Доступные форматы

Modellingtheimpactofproposedpoliciesonpubs

potential impact of a mandatory freeoftie option and the impact of introducing a

The additional introduction of a mandatory freeoftie option, which is a more rigorous

Contract management costs may be a fixed cost, and therefore constant

Dryrent increases by a proportion of the increase in the pubs divisible balance

Option 1 Option 1 Option 2 Option 2

Option 1 Option 1 Option 2 Option 2

Option 1 Option 1 Option 2 Option 2

Вам также может понравиться