Regional Development and Trade Theories

Document Sample
Regional Development and Trade Theories Powered By Docstoc
					Lecture 19
Housing as an economic good

o    Two ways of viewing housing
    1. Flow: Value a stream of services—equated to rent

    2.   Stock: Investment, an asset—equivalent to purchase price

o     Housing is a merit good—private in that it is excludable and
      rival in terms of the way it is consumed
    o     Its consumption is encouraged as a matter of public

    o    Viewed as a way of establishing social capital

    o    Investing in it is encouraged (tax deduction for mortgage

                    Edward W. (Ned) Hill, UST 605
Rental equivalency

o Value of house is roughly equal to the stream of net rents
  from the property

     SP = R0 + R1/(1+r) + [R2/(1+r)2] + [R3/(1+r)3] + … + [Rn/(1+r)n]
     Where r is the discount rate and R is the net rent (rental
     payment less operating costs)
     SP = Σ R/(1+r)t

o Sales price exists in a range that is influenced by bargaining
  ability, information, psychology, and family, locational, and
  financial pressures

                       Edward W. (Ned) Hill, UST 605
Some unresolved public policy issues about housing--

o Encourage a nation of homeowners for reasons of social
  stability, neighborhood stability, and wealth-building through
  asset accumulation

o Where is the limit?

o Can government do people and families harm by pushing the
  homeownership rate too high?

                   Edward W. (Ned) Hill, UST 605
                     Homeownership Rates in the US 1980(1) to 2009(1)




                                                                              63.8% Edward W. (Ned) Hill, UST 605                       5
ex.html, retrieved May 3, 2009
                       Home equity declines from a peak of $12.5 trillion
                       in 2005(4) to $7.8 trillion in 2008(4)---a 37% decline

                                                                        Peak 2005(4)
                                                                        $12.5 trillion

                                                                                         2008(4): $7.8

Source: Board of Governors of the Federal Reserve                                                        6
                                                    Edward (Ned) Hill
System, Flow of Funds Tables, B100
                       As home values decline, mortgage liabilities as a percent of
                       asset value climbs Quarterly 2002(4) 2008(4)



Source: Board of Governors of the Federal Reserve                                           7
                                                    Edward (Ned) Hill
System, Flow of funds Tables, B100
                       In Fourth Quarter 2008 Case-Schiller National Home Price Index
                       declined by 50.8 points, lost half of its gain from 2000 and declined
                       by 26.7% from peak

                                                                                                 Peak 2006(2)


Source: Standard & Poor’s Case-Schiller Home Price Index 2008(4)   Edward (Ned) Hill
                                                                                       Base value = January 2000 retrieved March 22,2009
                     Housing prices still declining; Sunbelt hit hardest
                     Speculative overbuilding in Sunbelt; Low demand in Frostbelt

Source: Standard & Poor’s Case-Schiller Index, 2008(4) )     Edward (Ned) Hill retrieved March 22,2009
                      New housing starts plummet; slide starts in January
                      2006 (thousands of units)

Source: St. Louis Federal Reserve Bank, FRED,                                                                     10
                                                                  Edward (Ned) Hill   Monthly data 1/79 to 2/09
                     Mortgage default rates are climbing
                     National Average at 3.5%; Florida is at nearly 8%

Source: TransUnion,                                                            11 retrieved   Edward (Ned) Hill
March 22,2009
 Homeownership rates by race 1994 to 2003

                      1994       1995       1996/1       1997      1998   1999   2000   2001   2002   2002/r        2003
U.S.                  64.0       64.7         65.4       65.7      66.3   66.8   67.4   67.8   67.9    67.9         68.3
Non-Hispanic White    70.0       70.9         71.7       72.0      72.6   73.2   73.8   74.3   74.5    74.7         75.4
Black, total          42.3       42.7         44.1       44.8      45.6   46.3   47.2   47.7   47.3    47.4         48.1
Hispanic or Latino    41.2       42.1         42.8       43.3      44.7   45.5   46.3   47.3   48.2    47.0         46.7
Housing Vacancy Survey, Annual Statistics, US Bureau of the Census

                                         Edward W. (Ned) Hill, UST 605                                         12
Four other issues
o How do you target federal housing resources?
   o Tax expenditures through mortgage dedication
    o   Direct, targeted spending
         o Supply-side interventions (construction)
         o Demand-side interventions (income supports)

o How large should the federal involvement in providing housing
  to the poor?

o What is the role of existing stock in supply-side housing
  policies (versus new construction)?

o What is the role for demand-side approaches (vouchers)?

                     Edward W. (Ned) Hill, UST 605
Fundamentals of Housing Economics
Flow: Market for housing services
Consumers choose between bundles of housing services consist
o The direct flow of housing services from the house or
  apartment itself (hard to measure but conceptually sound—
  number of bedrooms, baths, quality of the structure, etc.)

o Amenity bundle of the neighborhood

o Public services embodied in the address

o Access characteristics of the location

o Tax cost, maintenance cost, rental rate (or monthly
  mortgage cost)

o It’s a Tiebout world

                   Edward W. (Ned) Hill, UST 605
Housing demand
o Tenure choice
o Household formation
   o Immigration
   o New households
o Life-cycle hypothesis
o Price & ability to finance
o Price elasticities owner-occupied housing (Edwards)
   o Italy     -0.09
   o Japan     -0.73
   o US        -0.32 to -0.51
o Income elasticities range from 0.31 to 0.65 (normal good)
o Variance in income also is a deterant: 10% increase in
  variability (sigma) reduced homeownership by the same
  amount as a 5% decrease in income
o Tax policy—Federal Income Tax: 1940 40% of dwellings owner-
  occupied; 1999 67% are owner occupied.

                   Edward W. (Ned) Hill, UST 605
Who rents; totals to 100%
Varady & Liipman (1994)

o Families: 17%

o Lifestyle renters: 21%

o College graduates starting out: 26%

o Black renters: 15%

o Elderly life cycle renters: 10%

o Struggling blue collar renters: 11%

                   Edward W. (Ned) Hill, UST 605

o What would a forecasting equation look like?
   o Estimated the size or change in the market?
       o Aggregate Supply (construction/materials costs, financing
         rates, land prices,
       o Aggregate Demand (price, after-tax real income, household
         formation rate)
       o Reduced form equations focus on price

   o Examine the characteristics of demand for a household

   o Estimate the value of the unit (Hedonic model)
       o Dependent variable: cost or value
       o Independent: size, rooms & types of rooms & features, age,
         quality, existence & location for amenities & disamenities,
         public services, zoning restrictions & zoning benefits
       o Schools as a special case
                   Edward W. (Ned) Hill, UST 605
Housing as a capital market problem
o Purchase is lumpy—rare with large transactions costs (search,
  closing, moving, & psychic costs)

o Investment is difficult to liquidate—therefore there is risk
  aversion on the part of buyers—because the markets are “thin”
  (few buyers and sellers in the market at any one time—
  especially by quality segments)

o Investment is risky

o Purchase needs to be financed (reason why in US can evict,
  causes problems in other nations when you cannot evict—
  increase lender risk)

                   Edward W. (Ned) Hill, UST 605
Ed Mills’ Critique

o US is over-invested in housing (compared to other capital
  assets) as a demander of capital due to:

   o   Mortgage interest rate deduction

   o    Lack of equality with rental payments (we do not charge
       homeowners an imputed rent and tax it as income like with do
       with landlords)

                    Edward W. (Ned) Hill, UST 605
How imperfect is the housing market?
A perfect market requires:
o Large number of buyers & sellers
o Large number of providers of new units (small relative to the
   market with ease of exit & entry). Some markets the issue lies
   in the control of land. Reason why the US has benefited from
   large numbers of small landholders
o Homogeneous good (Edwards adds this, I disagree; in practice)
o Perfect information
o Constraints on the supply of housing—government regulation in
   both the housing market (quality), the land markets (zoning),
   and in law itself (the primacy of property rights and its impact
   on negative externalities).
o Housing embodies many goods both stocks & flows—public
   services, housing services, social capital, investment
o Despite this in the US it is an efficient market in the Tiebout
                   Edward W. (Ned) Hill, UST 605
The uncertainty of housing prices around a point.
Note the distribution and “thick” curves

                                               o The price is a distribution of
                                               o Mean is the equilibrium for a
                                               submarket (note difference in
                                               language from Edwards)
                                               o “Thick” supply & demand
                                               curves. Right shows a + 10%
                                               o Reflect imperfections in
                                               information, imperfect
                                               substitutes or cross-price
                                               elasticities between
                                               o Differences in bargaining
                                               ability bargaining experience

               Edward W. (Ned) Hill, UST 605
Rent control—a price ceiling

               Edward W. (Ned) Hill, UST 605
Housing vouchers; housing as a normal good and
then as an inferior good

              Edward W. (Ned) Hill, UST 605
Six complementary models of the housing
Six complementary types of housing models
1.Trade-off Models (Access for space)
Bid-rent models covered last week

      Total housing cost
SR1                                               Site rent

                                                  Transportation cost to the CBD ( T = td)
                                                  Housing cost excluding site rent

      O or CBD               d1                                    E      Distance form
                                                                          CBD d

HC housing cost; TC transport cost, SR site rent, E Edge of
metro area
Graph of site rent is inverse of transport cost graph

                           Edward W. (Ned) Hill, UST 605
Relationship between transport cost, site rent,
building cost, and density


                                                                   cost to
              Site rent
                                                                 Housing rent

 D0   DB            DE                 O                  B   Edge of metro area

 MC of building related to density
 D is density at distance from CBD or O
                          Edward W. (Ned) Hill, UST 605
Six models
2. Cultural agglomeration models
o   People sort themselves out by income and desired amenity
    packages into socially homogenous neighborhoods

o   Leads to social segregation (unless diversity is viewed as an

                  Edward W. (Ned) Hill, UST 605
Six models
3. Tiebout model of service provision

o   Relationship between government services, tax cost, and
    housing prices in making locational decisions

o   Model requires
    o   Many small jurisdictions (government units or neighborhoods)

    o   Each with different amenity packages & tax costs

    o   Full mobility on part of potential residents, who have diverse
        preferences in terms of public amenities

o   People vote with their feet to determine proper mix of
    public amenities and tax costs

                    Edward W. (Ned) Hill, UST 605
Six Models
4. Hill-Bier Neighborhood demand model
o   Explains movement in housing values as reflection of
    the regional demand for the human capital embodied
    in the neighborhood

o   Neighborhood characteristics are effected by regional
    macroeconomic forces

                Edward W. (Ned) Hill, UST 605
The fifth model—filtering
Filtering—Housing as Newtonian Mechanics

Welfare filtering—people

Housing filtering--units
Filtering model of housing
o Confusion about filtering model:
   o   Does the model focus on housing units or households?
   o   What is the welfare question?

o How does filtering theory influence policy?

                    Edward W. (Ned) Hill, UST 605
The classic filtering model
o Racliff (1949):
   o Housing tends to move downward in the quality and value scales
      as it ages

o Define filtering as:
    o Changing of occupancy as the housing that is occupied by one
      income group becomes available to the next lower income group
      as a result of declines in the market

o Still unclear: units or people
   o Focus of post-WWII policy: units

                    Edward W. (Ned) Hill, UST 605
Filtering: Process or results

o Is it an allocation process where households change their
  housing to match their changing incomes and preferences?

o Larry Bourne talks about “welfare filtering” where poor
  households improve their housing quality over time

                   Edward W. (Ned) Hill, UST 605
Assumptions behind the model

o   A metro area is the market

o   Units are graded by quality and there exits a hierarchy of

o   Hierarchy of units is correlated with households income

o   All households of equal income wealth have equal access to
    the housing stock

                   Edward W. (Ned) Hill, UST 605
Spatial assumptions
1.   New construction tends to be superior to existing

2.   New construction tends to be located on the periphery

3.   New construction rate > population growth rate

                   Edward W. (Ned) Hill, UST 605

o   “Excess” number of units exerts downward pressure on
    existing rents and prices

o   While quality may decline it does not drop as fast as
    market value

o   Stock can be removed, and the maximum amount of stock
    to be removed is equal to excess supply

                   Edward W. (Ned) Hill, UST 605

o For welfare filtering to exist poor households must move up
  the housing hierarchy faster than quality declines due to
  deferred maintenance

o Implies that housing is becoming relatively cheaper

o Policy conclusion: Produce a surplus of new units at the
  middle and upper ends of the distribution

                  Edward W. (Ned) Hill, UST 605

o Relevant comparison should be between the real incomes
  of out-movers when they moved into the unit and of the
  family moving into the unit

o Not the current income of the out-mover and the in-mover

o The realities of the income distribution implies that the
  market cannot introduce enough housing at the top of the
  housing spectrum to satisfy all of the demand for housing
  services from low income families (there are more middle
  and low income families than upper income families)

o Discrimination impedes the filtering process

o No differentiation is made in the models between the
  behaviors of owners & renters

                   Edward W. (Ned) Hill, UST 605
Policy complaints
o Impact of filtering on maintenance—leads to deterioration and
  has spill-over effects into neighborhoods

o Local government impedes filtering through land controls

o Stock removal not done quickly enough
   o Again existence of externalities

                    Edward W. (Ned) Hill, UST 605
Empirical evidence
o Filtering chains were shorter that expected
   o longest about 4 moves

o Filtering chains depend on:
   o the regional housing market

   o state of the economy

   o original value of unit

   o factors that end the chain
       Demolition, Move outside of the region

                    Edward W. (Ned) Hill, UST 605
Race and filtering
o Marullo:
   o Filtering chains restricted to narrow stratum of income

   o Filtering works for the well-to-do often breaks down for poorer
     households, especially black households

o In sum there is an argument for demand side housing policies

o Leads us to a housing adjustment model, which is Galster and

                    Edward W. (Ned) Hill, UST 605
Model Six
Galster-Rothenberg Model of the Housing

Key: Housing market is an interconnected array of submarkets
segmented by quality
Housing quality submarkets:
Part 1
o Housing is spatially fixed, heterogeneous, hard to change

o Housing units are aggregated into is a set of segmented and
  interconnected quality housing submarkets

o A submarket is a set of units that are close substitutes.
    o What goes into each submarket is revealed by the preferences
      of the demand side of the market (housing shoppers) and the
      tradeoffs they make.

    o Units in a submarket are not necessarily spatially contiguous
      but are viewed as reasonable substitutes for each other.

                     Edward W. (Ned) Hill, UST 605
Housing Quality Submarkets: Part 2

o Can adjust supply and demand within a submarket through
  conversions of existing units or through new construction

o Submarkets are inter-related; the ability to substitute
  successfully becomes increasingly difficult the further away
  you get from each other in terms of quality

                   Edward W. (Ned) Hill, UST 605
Housing Quality Submarkets: Part 3

o Three sets of actors
    o Households: demand
    o Builders: supply
    o Converters: supply

Key to understanding the relationship between submarkets:
o Qx quantity in submarket x;

o px is price of unit in submarket x

o Own-price elasticity within a given submarket x:
      ηQX = (%D QX) / (%D PX)

o Cross-price elasticity between two submarkets x and x+1:
       η = (%D QX) / (%D PX+1)

                   Edward W. (Ned) Hill, UST 605
Why are elasticities important?

o The greater the degree of substitutability between two
  submarkets the higher is the cross-price elasticity

o Differences in cross-price elasticity reason for the non-
  uniform response to demand shifts among the various

                   Edward W. (Ned) Hill, UST 605
Two market periods
Short & Long runs

o Short Run: Time is so short that cannot convert housing
  stock or build new stock.
   o Supply on market determined by reservation prices of current

   o Owners of rental units examine market value (MV)

   o Owner-occupied units compare MV-Moving costs

o Medium Run: Supply responses are relevant
   o Reservation prices determines behavior of existing stock

   o Conversion or existing stock and new construction gives
     elasticity to the supply function

                    Edward W. (Ned) Hill, UST 605
The demand side of the market
Demand Part 1

o Household selects a quality submarket such that maximize
  Utility (H, X) subject to: Y,W
      where: H = housing consumption
                X = other goods
                Y = current income
                W = wealth
o Traditional demand curve, where Qx is the quantity of units
  available in quality submarket x

$ MVX                      Market ValueX (MVx) = f(QX)

                                                     <   0

                              DX                    δ = change
                                     QX DWELLINGS
                   Edward W. (Ned) Hill, UST 605
Demand Part 2

o The own-price elasticity is related to the quality level in that
       o i.e. the poor have less choice therefore their demand is less price
       $       S2                              Demand shifts.
  p2                                                     What are the supply
                                                         responses if supply if
                                S1                       perfectly elastic? Perfectly
                         D1 D2
             q1     q2                 Q

o Cross-price elasticities are important; substitute goods are highly
          cross-price elastic
o S1 is perfectly elastic—a small change in price brings forward a huge
          (infinite) supply response (p1, q1) to (p1, q2)
oS2 is perfectly inelastic—a huge price increase brings forward no
          additional supply (p1, q1) to (p2, q1)
                         Edward W. (Ned) Hill, UST 605
Last (repetitive) word on elasticities

o Cross-price elasticities are inversely related to the quality

o Larger quality differentials results in lower cross-price
  elasticities because the other submarket is a poor substitute

                    Edward W. (Ned) Hill, UST 605
Demand for dwellings

MVX = f(Y, T, PX, MVX+i)

o    Causal factors that are exogenous to the housing market:
    o Income (Y),
    o preferences(T),
    o price of non-housing goods (PX)

o    Causal factor that is endogenous to the housing market
    o MVX+i

                    Edward W. (Ned) Hill, UST 605
The supply side of the market
Short run supply function (Sx) in one submarket

In the short run you cannot add to supply Q is quantity of dwelling units in
                                                       quality submarket x

                                                       MV’ - minimum reservations price
                          SX                           MV2-price inflection point. No more
     $MVX                                              physical supply can enter the
                                                       market. You have a pure auction
                                                       market with no supply response
                                                       after this point—perfectly inelastic
   MV2                                                 supply response
                                                       Qmin - number of units held by
   MV’                                                 owners in the quality submarket
                                                       with the lowest reservation price
                                                       Qmax - preexisting submarket
                                                       stock (At start of the market
            Qmin       Qmax               QX           period)
                                                       Note: This part of supply curve is
                                                       perfectly inelastic
                                                       Slope is a function of the
                                                       of the reservation prices of owners

                       Edward W. (Ned) Hill, UST 605
Market period: Equilibrium

o Specifies number of dwellings occupied: Q*
o Specifies number of vacant units: Qmax - Q*
o Max Stock (Qmax) and Min Stock (Qmin) together define the
  market period supply function



   MV*                                                     Qmax - Q*
                                        Vacancy rate =
   MV’                             DX                           Qmax

                                                Edward W. (Ned) Hill, UST 605
               Qmin   Q*   Qmax

                                          Qx                                    56
           Market periods and multiple submarket supply

Short run supply functions in three
submarket x1, x2, x3 and Smed is medium              This graph repeats the previous graphs for
term supply function across the submarkets           multiple submarkets and Qx represents the
                                                     quantity in each submarket
Note: Difference in price elasticity in SR
and MR                                               Shows minimum stock, array of reservation
                                                     prices & maximum stock define the market-
                                                     period supply function for each submarket
                                                     Upward slope of Sx1 is determined by the
                                                     distribution of reservation prices across
                                                     owners in each submarket
                                                     Next slide treats these three markets as three
                                                     market periods and shows movement from
                                                     short run to the medium run.

                                     Edward W. (Ned) Hill, UST 605
From short run to medium run in one quality

                               Start (Q*, MV*)induces conversion of units into
                               market x
                               Shift in supply curve from Sx1 to Sx2

                               New equilibrium (Q2,MV2) and maximum number
                               units to Q2max Note that the price fell due to
                               additional supply and no shift in the demand curve.

                                  Now assume a demand shock due to immigration
                                  Start with equilibrium (Q2,MV2)

                                  Dx1 shifts to Dx2, equilibrium (Q2max, MV3,)
                                  Induces a supply response Sx3 and (Q4, MV4)
                                  The medium run supply curve (function) is the
                                  path through the stable equilibria and beginning
                                  with Q1min Follow the red arrows

               Edward W. (Ned) Hill, UST 605
Medium run supply adjustments

o Medium run is the required to adjust the stock of housing by
  way of new construction, quality adjustments, or stock

o Suppliers maximize their risk adjusted rate of return

o Must consider alternative (non-housing) investments and net
  return on quality adjusted housing

   o That’s why opportunity cost of capital is part of the cost

                   Edward W. (Ned) Hill, UST 605
Moving between quality submarkets From Q
(quantity in submarket x) on the x-axis to x on the
x-axis (x indicates a particular submarket)
New construction: The builder’s decision
o Consider benefits and costs in each submarket
    Costs (C) include land, construction costs, PV of maintenance
     in submarket x, and opportunity cost of capital
    Benefits (MV): discounted PV of expected sales (or net rental

o Both benefit and cost functions are assumed to be
  monotonic (and possibly non-linear) functions of quality

       dB                                           dC
          >0                                              >0
       dQX                                          dQX
                    Edward W. (Ned) Hill, UST 605
 New construction in submarket 2

 Goal: Max (MV - C)



 C = New construction cost function
 MV = Total revenue or market value function
 Xi = particular quality level within housing quality submarket X

                       Edward W. (Ned) Hill, UST 605
               Housing quality submarket

                                                New Construction Cost Curve Cx

                  $MV Costs
                                         C2                                   C3

                  X2 X2*               X2’              XNC             X3*   X3
* Indicates the optimal location
other locations are beginning points
                                        Edward W. (Ned) Hill, UST 605
Housing submarket quality
o MV : MV possibilities submarket Xi
o X3 target submarket for new construction as
  (MV - NC) max
   o NC = new construction cost function

o Conversion: conversion function CX shows the movement from
  where you start to the new optimum X*

o Note decreasing return in MV function (the slope of the line
  flattens) and increasing marginal cost at end of NC function

                   Edward W. (Ned) Hill, UST 605
Conversion: Medium run supply behavior
Rate of return maximization for new construction
and conversion
                                  Case of owners in 4 submarkets
                                  New construction function is C,
                                  includes opportunity cost of capital,
                                  maintenance, and up or down
                                  conversion for each submarket
                                  For example:
                                   X2 should upgrade to X2’
                                   X4 should downgrade to X4’
                                   X3 is fine where it is
                                   X1 cost of maintenance exceeds
                                            revenue (C1 > MV) so the unit
                                            should be abandoned

               Edward W. (Ned) Hill, UST 605
Why do the conversion functions slopes look the
way they do?

o Conversion costs increase as span between the originating
  quality submarket and the target quality submarket widens

o Generally it is more expensive to convert older dwellings
  then to build new

                   Edward W. (Ned) Hill, UST 605
Conversion options: Part 1

1.   Maintain position: Consider PV of maintenance and
     opportunity cost of embodied capital

2.   Upgrade: conversion cost, PV maintenance and
     opportunity cost

3.   Downgrade: Passive maintenance or subdivision
4.   Nonresidential use

5.   Abandon: opportunity cost of embodied capital is worth
     more invested in other assets.

                   Edward W. (Ned) Hill, UST 605
Conversion options: Part 2

o Note the role of the opportunity cost of embodied capital

o in C function it is incorporated therefore where MV > C you are
  sure that rate of return is superior to other investments

                   Edward W. (Ned) Hill, UST 605
Consequences: Part 1
1.   Conversion into any specific submarket depends on the
     characteristics of the standing stock in the other

2.   A unit can move to just about any other submarket
     o   However new construction is much more constrained
     o   Can’t profitably meet demand in low quality markets

3.   Cost of conversion and the initial quality configuration of
     the stock influences the aggregate supply response to a
     given submarket

                   Edward W. (Ned) Hill, UST 605
Consequences: Part 2

4.   Conversion tends to dominate lower quality submarkets
     o   Downgrading relatively inexpensive way to meet demand
     o   Housing codes and zoning laws increase difficulty to build new
         housing at these valuations

5.   New construction dominates higher quality
     o  Submarkets generally more expensive to upgrade due to
        new technologies

                   Edward W. (Ned) Hill, UST 605
What happens with high quality cost

                            Two owners with cost functions:
                            C3 and C3*
                            New units in X3 have much lower costs
                            Slip in revenue function to MV1’
                            Profits erode for C3 They shift supply into
                            market X2 and have conversion function
                            Ripple effect in market X1 results in C1* >
                            MV1* and abandonment

            Edward W. (Ned) Hill, UST 605
              Aggregate impacts on three rental
              submarkets—low, medium and high

1) Original demand & supply           6) Depressed rents in H causes       9) Lower rents in M send
functions subscript 1                 expansion in supply in M so that     supply to L SL6 (light blue)
                                      SM1 to SM4 (blue arrow)
2) Shock—new construction in high
market shift SH1 to SH2 (red arrow)   7) Results in SH2 shifting back to
                                                                           10) Surplus supply causes
                                      SH4 and rents in M fall to MVM4
3) Rents go to MVH2 (red arrow)                                            some abandonment and a
                                      8) Demand shifts into M from L       second shift to SL7 where
4) Renters move from M to H and QH2
                                      result in DL5                        changes dampen out
is consumed
5) Demand curve shifts in M to DM3
                                       Edward W. (Ned) Hill, UST 605