RAHUL DEODHAR Rahul Deodhar The idea‐book discusses the principles that determine firstly how cities evolve over seven phases. Secondly, we see how Affinity Factor model may help understand how development spreads or distributes within the city. We observe how these principles impact selection of office location and how we can predict the future of developing business districts. We try to understand how house prices get influenced. Finally, based on the principles discussed, we try to work out a possible township model. Oct 09 How Cities Develop? Idea‐Book on
The idea-book discusses the principles that determine firstly how cities evolve over seven phases. Secondly, we see how Affinity Factor model may help understand how development spreads or distributes within the city. We observe how these principles impact selection of office location and how we can predict the future of developing business districts. We try to understand how house prices get influenced. Finally, based on the principles discussed, we try to work out a possible township model.
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
R A H U L D E O D H A R
RahulDeodharThe idea‐bookdiscusses theprinciples thatdetermine firstlyhowcitiesevolveoversevenphases. Secondly, we see how Affinity Factor model may help understand howdevelopmentspreadsordistributeswithinthecity.Weobservehowtheseprinciplesimpactselection of office location and how we can predict the future of developing businessdistricts. We try to understand how house prices get influenced. Finally, based on theprinciplesdiscussed,wetrytoworkoutapossibletownshipmodel.
Realestatedevelopment ineverycity isunique.Stillhiddenwithin,arecertainprinciples that are common. To understand it, we need to understand twocentral concepts. First, how town evolve and second how evolution happenswithinatown.
I propose a seven phase model explaining how a population surrounding abusinessorfactorytransformsintoatown.Throughthetransformationwepointtosomeimportantdevelopmentsintermsofpeopleandtheirwork.
The idea book postulates a growth model called “Affinity Factor Model” toexplain how localities develop within a town. “Affinity factors” are those thatdrive the citizens towards them – e.g. business district and schools are keyaffinityfactor.
Themodels help us understandwhy airports, usually built outside city limits,attract residential populations. Or, on a lighter note, we can guess where acompanywilllocateitsoffice!
The work can be shared for noncommercial use through proper attribution asexplained in Creative Commons AttributionNoncommercialShare Alike 3.0UnportedLicense
PhaseI) SeedPhase:Inthisphasetheseedofdevelopmentissown.Thisistypicallyabusinessdistrictorfactoryisestablished.Thebusinessdistrict is plannedwith a certain population inmind. Thepeoplewhoworkherelivehere.
PhaseII) Development of Support infrastructure: In this phase supportinfrastructuredevelops.Thisleadstoslightincreaseinpopulation.Generally, planners already account for these. This typicallyincludes:
Retailers for regular goods (groceries, pharmacy stores,gasstationsetc)
PhaseIII) Business Expansion: The dominant businesses attract othersupporting industries and a factory settlement starts becoming atown.
This leads to further population expansion but thepopulationstilllivescloser.
PhaseIV) Strengthening of support infrastructure: In this phase thesupport infrastructure itself becomes an income generatingactivity. The breadth and depth of services increases drasticallyleadingimprovementinqualityoflife.Typically,
PhaseVI) Super straining: In this phase municipal innovations indebottlenecking, infrastructure additions create some relief.Though slums increase and cost of doing business startsskyrocketing.
AffinityFactorModelTherearesomecharacteristicsthatattractustoanylocalitywithinacity.Suchpreferences lead to clustering of similar people – though they never meet orinteract with each other. Affinity factor model postulates a basic frameworkthroughwhichwecanaggregatethesefactors.
1. Affinityfactorisfactorthatattractspeopletoaresidentiallocation.a. Affinity factors exert a force that can be expressed similar to
gravitationalforce.b. The force of attraction between an Affinity factor and a locality
(neighbourhood) is directly proportional to importance of factorand inversely proportional to the square of distance betweenthem.
c. Sincethereisnodocumentationandcalculationofaffinityfactors,we cannot surely say if we should use distance or square ofdistanceasinNewton’slawofgravitation.
d. The ultimate preference of location is a vector sum of all theattractionforcesactingonthelocality.
2. Theevaluatorrelativelysetsimportance.a. A single worker tends to set higher weight for proximity to
4. Thedistancevariableisinterpreteda. Based on certainty of commute time: A 30‐min drive (average
time) through safe lonely roads is preferred over 30‐min drivethroughmostlycrowdedroads.
b. Basedon safety:A45‐mindrive throughabsolutely safe roads ispreferredover20mindrivethroughdisturbedneighbourhoods.
DispersiondevelopmentandSubsequentAffinityfactordevelopmentDispersion development is the understanding of how population settles giventhe Affinity factors existence. Subsequent Affinity factor development is afunction of the existing and planned settlement. These two phenomenacompliment each other iteratively. The overall development is thus fractal innature.
DispersionDevelopmenthelpsusunderstandwhichareaswill seehousepricerise. It can help predict median prices in a locality. It can definitely predictrelativepriceranksbetweenlocalitiesorneighbourhoods.
Subsequentaffinity factordevelopment isdependantondispersionat thetime.Thishelpsunderstandpracticalquestions.Wecanpredictwhereacompanywillrelocate its office.We canpredict if newbusinessdistrictwill be successful ornot.Wecanevendesignstrategytomakeitasuccess.
DispersiondevelopmentTheAffinityfactormodeldriveshowthedispersionofthecityoccurs.Dispersionforces radiateoutof affinity factor like, businessdistricts (orworkplaces) thatare first affinity factor. Such dispersion forms areas based on commutes. Theinnermostcircle,representingshortestcommutes,developsfirst.
SingleAffinity‐FactordispersionWe can see a good example of such one‐factorinfluenced development at industrial townships orsettlements around as single manufacturing plant.Herecommutesareoftenaslowas10mins.
Thedevelopmentstartsclosertofactorygatesasthisminimizes commute time (even in this small scale).Development eventuallymoves outward gradually ina circular fashion. The concentric circles representcommutes.
Nowifweaddaconnection,sayaroadormetrolinkthenweinfluencethedispersion.Dispersionaround the Affinity Factor is higher along thisconnection as commutes are easier along theroad or metro lines. The dispersion is nowskewedalongtheroadormetroline.Theshadedarearepresentsthenewdispersion.
However, rarelydowehavesuchsingle factorexamples in real life.Usuallyasadditional factors get introduced we start getting skewed distributions. Fullyformedcitiesareexamplesofmultipleaffinity‐factordrivensettlements.
MultipleAffinityfactordevelopmentWe can therefore extrapolate thedispersion in multifactor localities. Asmentioned, the Affinity factors forcesare vector additions and variouscombinations can be worked uponbased on type of Affinity factor andcommutes.
Alongsidewehaveshownanexampleof3 factordispersion (shaded)withmainroad (arterial connection) and ananciallary road (e.g. a side road). The scheme is indicative and notmathematicallymodelled.
SubsequentAffinityFactorDevelopmentJust as Affinity factors influence dispersion, dispersion also influences newdevelopments.ThisrefertodevelopmentofnewbusinessdistrictsorexpansionofexistingbusinessdistrictsordevelopmentofotherAffinityfactors.Thismeansbasedon current location (homeandworkplace)ofpopulationwe canpredictwhatareasaremorelikelytobethenextbusinessdistricts.
There exists betweenAffinity Factors and dispersion an interdependence. Thefractalnature(iterativewithsimplerules)ofdevelopmentpossiblycausesthis.
The interdependence is breakable and initiating a new Affinity Factor usuallycreatesforcesofdistortion.Thisnewaffinityfactorhastobeahighimportancefactorandcannotsimplybeaparkorgarden.Usually,newairport,newbusinessdistrict(CanaryWharfe.g.)hasthepotential.Still,suchnewfactorstakelongertopay‐backforinvestors.
ImpactofzoningandotherregulationsTheargumentsandideasaboveareessentiallyforanorganicallydevelopingcity.Zoning directs or channelizes the development but overall organic natureremains.Sincedevelopmentisiterative,a5‐yearzoninglimitation(e.g.)willalterthe cities development course forever though its influence wanes with time.Hencewhenwearelookingatacityanditsfuturedevelopment,itisimportanttoknowthehistoryaswell.
SomepracticalobservationsandinsightsNow let us use the concepts above and distil them into practical applications.These should help real estate brokers, developers, investors and users tounderstanddevelopmentbetter.Ihaveincludedpriceunderstandingseparately.Iwouldlovetohearexamplesreinforcingorcontradictingtheseobservations.
Wherewillanewofficebelocated?Imagineacompanythathastoshiftitsoffice.Thenewoffice,ideally,shouldbelocatedsothatitisconvenientforemployees,customersandsupplierstoreach.Soitfollowsthatifweconstructimportanceandspreadofemployees,customersand supplierswe can find the optimum location. This gives us a neat logic forwhy businesses often seen clustered around a location. So we can infer thefollowing:
1. The importanceof topmanagement residentialdispersion ishigherandinsomecasesitisonlythingthatmatters.
• Theoffice location ismostly themostconvenient location for topmanagement(orkeydecisionmakers).
• This results in most offices locating closer to prestigiousresidential areas resulting in longer commutes for most of theemployees.
SettingupnewbusinessdistrictsSuccessofnewbusinessdistrictsisdefinedbyconvenienceofcommutefortopmanagement. The top management prefer to stay in prestigious residentialareas;sonewbusinessdistrictmustbeaccessiblefromsuchareas.Therefore,wecaninferthefollowing:
1. If it takesmore time to reach thenewbusinessdistrict then its successoddsarelower.
• The travel time ismeasured fromkeydecisionmaker residentialareasandemployeeresidentialareas
2. If a new business district locates on the connection between old,established business district and key residential areas then it is morelikelytobeaccepted.
• Even here it has to offer lower rentals, higher floor plates,additional conveniences likemore parking per seat,more visitorparkingetc.
3. If connection to the newbusiness district runs through congested, low‐incomeareastheoddsofsuccessdecrease.
• Fortunes of ailing business districts can drastically change bydeveloping new connections (roads, rails, metro etc.) to topresidentialareas.
InfluenceofairportsNewairportsareoftenlocatedoutsidethecitieswheretheaircraftnoisewillnotdisturbcitizens.Yetcuriously,theresidentialareaseventuallycomeupclosetoairports.Thereisareason.Theadministrationusuallybuildsafast‐lanehighwayor high‐speed train to the airport directly frombusiness district. This cuts thecommutetimesignificantly.Soonwefindthatitiseasierandfastertocommutefrom airport than our congested residential neighbourhood. Naturally theresidentialdevelopmentmovesclosertoairport.
1. If the connection is a metro then it makes sense to buy land near themetro stations. Initially the high‐speed train has no stops in between.However eventually normal metro trains run along that line and thosestopatin‐betweenareas.
2. If the connection is a road then red‐lights or signals are best place tomakeinvestments.Signalsonthisroadindicateimportanceandthereforeeasierexit.
3. If there are two connections then land between them and towards theairportisofprimesignificance.
4. Logically itmay appear that near‐airport lands are best used for hotelsand other tourist infrastructure. But there is lot of residentialdevelopmentaswellanditisnotlimitedtolow‐incomehousing.
5. The above is general organic development and zoning or otherregulationsmaypreventoralterit.
DeterminingRealEstateprices&developmentUnderstanding price is very difficult exercise. The complexity is result of twintracksuponwhichpricedepends.Pricevariesspatiallyandovertime.
Inmyexperience it isbetter tounderstand fundamentalpricevariationacrossthecity.Thenweneedtounderstandhowfundamentalpriceschangeovertime.And finally, we superimpose adjustments for Real Estate industry cycle andeconomiccycles.
SpatialdistributionSpatialdistributioniseasilydeterminedusingAffinityFactormodel.TheAffinityFactor model gives us lines of influence adding at a location. This is a vectoraddition implies the resulthasavalueanddirection.Thevaluecanbeused tounderstandrelativepricesacrosslocalities/neighbourhoods.Thedirectiontellsus,indicatively,whatAffinityFactorismostinfluentialandhencewhatcouldbefundamental price level. The relative rankingofneighbourhoods is constant inshorttermandchangesonlygraduallyoverdecades.
Letusrevisitthemultipleaffinityfactordiagram.HerethepricesinproximityofAffinity factor 1 (say industrial park)will be determined by wages inindustrial park. Similarly prices nearAffinity factor 2 (say IT park) will beinfluenced by IT salaries. Anecdotally,thefundamentalpricesnearITparkwillbe higher than those around Industrialpark.
However,thepricesatthecentralintersectionwillderivefromallthreefactors.Further, if Affinity Factor 3 is a golf course residential community then pricesaroundthatwillbedrivenbyhighestincomeearnersamongstallthreefactors.
ChangestofundamentalpricesovertimeWenowneedtounderstandchangesinfundamentalpricesovertime.Thisisafunctionofmedianincomeinthebusiness district andneighbourhood. Typically themedianincomeandfundamentalpricefollowthepathindicatedinfigure 4. This depicts the pricechanges in single Affinity factormodel.
As the Affinity factor is formedand developed the medianincome drops initially then setson a growth path. The initial
drop happens asmigrant population comes in to stay around the factor. Thispopulation is dependant on residents and therefore has lesser income thanresidents. If township planning is done properly, new growth opportunitiesemergeleadingtogrowthinincomesandthereforefundamentalprices.
ActualPricemovementActual price is result of certain factors weighing in on the fundamental price.Fundamentalpricesareeasiertodefendevenindownturnsandformsomesortof floor for prices in the area. However any decision related to real estateinvestment must consider future actual prices. Following are the factors thataffectactualprices:
1. Loan to value ratio of banks: Banks give certain part of house value asloan. The rest amount comes from individual / household savings. Forsame down payment, changes in LTV impact affordable house pricedrastically.
• E.g. If down payment is $10,000 then at 90% LTV person canafford house of $100,000. But if bank changes down payment to80%thenaffordablehousepriceisjust$50,000.So10%changeinLTVcreateaffordabilityswingof50%.Thisdoesimpactprices.
2. Interest Rate Scenario: If people believe interest rates will continue toremainstableonlowersidethenhousepricestendtoincrease.
3. Policy intervention: Government can give tax breaks and incentives thatmayimpacttheprices.
4. Income profile changes: Overall income profiles may change as type ofbusiness in the business district changes. This is creeping change andtakeslongertime.
Following chart, figure 5, gives anexampleofactualpricesinalocalityover time. As the localityexperiences growth the pricesincrease. However as fundamentalprices taper off we see peak inactual prices and these correctthereafter.
3. TertiaryWorkers:Theysupportprimaryandsecondaryworkersaroundresidentialpremisesandbusinessdistrict.a. Theydohousekeepingatbusinessdistrictafteritcloses.b. They also critically support the primary and secondary workers
helpingthembeforetheygotoworkoraftertheyreturnfromwork.E.g. Metro train operators, airlines, house‐help, baby‐sitters, taxioperators,police,etc.
4. Dependant Population: This includes school (including high‐school)childrenandseniorcitizens.
• The usual material required at business district is transported before thebusinessdistrictopensorafteritcloses.• Secondary workers usually handle this activity. Same is the case for
• Foodandothertimecriticalmaterial(mail)movesintothebusinessdistrictduring the working time. The infrastructure needs to be planned for thismovement.Infrastructureimplies:• Parkingformailvans,foodvans,fooddeliverypeopleetc.• Loadingandunloadingbaysatofficesandshopsforabove
• Generallysomecapacityisrequiredformovingconstructionmachineryandmaterialsasthereisalwayssomeconstructiongoingon.• This includes utilities (power, water, gas, telephones) lines
ProposedTownshipmodelAbove ideas can be used to create an easy access township model. One suchmodelcouldbeasshownalongside.Wedrawarepresentativesegmentofatown–oftencalledsector.
1. The area within the circle iswalk‐able. The bigger circlesdenotelongerdistances.
2. Business district is big circle –probably distance covered bytaxiinthefirstmeterreadingor10‐mindrivetime.
lessmiddle‐incomegrouppeopleascomparedtoLIG.3. Dependencyratioisusuallylesserthanaverage.4. Car or personal vehicles infrastructure in addition to high frequency
publictransportation.a. Limited options for public transport are fine just the frequency
shouldbehighb. TypicallytoLIGandBD
HIGreferstorichpeopleresidentialarea.1. This is usually located around a leisure factor like beach‐fronts, lake‐
a. Additionally, high frequency public transportation is required asthereislotoftertiaryworkforcesupportingthisarea.
b. Directhighspeedhighwaytobusinessdistrictisrequiredc. PublictransportationisrequiredtoLIG(highfrequency)andMIG
(mediumtolowfrequencyisfine)
For bigger towns the sectors can bearrangedasbelow.Alternatively therecanbe multiple ways in which we can createtownshipwhilemaintaining theprinciplesdiscussedabove.
ThusIhavevalidatedtheseonlyanecdotally.Iwelcomesuggestionsandtestingof these principles and look forward to working on them. I will continue tochangemodifyoralterthetheoriesbasedonfurtherexperienceorresearch.
Users should exercise cautionwhile studying the principles. If in doubt [email protected]
AboutMeI worked as a buy‐side analyst with top hedge fund client of Morgan Stanley.Prior to this, I worked for CRISIL Research doing industry and companyresearch.Ihaveover8yearsofworkexperienceacrossvariousrolesstartingonthe shop floor to investment analysis. You can email me [email protected].