Get documents about "12207Paper APRIA 2007 - Madalasa V _ R Vaidyanathan
Document Sample
Home Ownership and Reverse Mortgage
As Components of Portfolio Choice in India
Madalasa Venkataraman
Fellow Program in Management
Indian Institute of Management
Bilekahalli, Bannerghatta Road,
Bangalore, India 560076
Venkatm03@iimb.ernet.in
R. Vaidyanathan
Professor of Finance & Control
Indian Institute of Management
Indian Institute of Management
Bilekahalli, Bannerghatta Road,
Bangalore, India 560076
vaidya@iimb.ernet.in
91-80-26993086
Introduction
Rational forward-looking agents ought to smooth consumption over the life cycle and
exhaust the asset stock accumulated during the working career in retirement. In the life
cycle-permanent income framework, total lifetime resources available to the economic
agent are an important determinant of consumption. Lifetime resources are comprised
both of the stream of labor income expected to receive over the remainder of the working
life and of assets owned by the consumer.
The portfolio of individuals about to retire consists of stocks, bonds, other financial and
real assets, such as housing. While stocks and bonds and financial assets are more or less
liquid and can be traded in the market, housing does not have a well-developed liquid
market, and housing incorporates both consumption and investment motives.
Real estate assets differ substantially from financial assets due to their dual role. Housing
wealth is demanded both as an investment good and as a durable consumption good.
While risk averseness in an individual might preclude him from holding financial assets,
the individual will still have to consume housing services (as in taking a house on rent)
and one way to hedge against inflation in rental values is to purchase the house itself.
Therefore, housing also has another role to play as a hedge against fluctuations in rental
value.
The lifecycle hypothesis of saving suggests that asset accumulation happens during the
early stages when there is an income inflow, and during the later years the elderly ‗live
off‘ the accumulated assets. However, housing, by its very nature, is inseparable into the
investment and the consumption components. The elderly would like to decumulate the
investment component, yet retain the consumption component of housing by living in the
same house. Given the virtually illiquid nature of housing and the high volatility of real
estate prices, the value of the housing investment is not guaranteed of a certain rate of
return. One such instrument that allows the elderly citizens to tap into housing equity
without selling the house and moving out is the reverse mortgage.
A reverse mortgage is primarily an instrument that allows the elderly homeowners
borrow against the accumulated home-equity without moving out of or being forced to
2
sell the home asset. Traditionally, reverse mortgages were designed as a product for
elderly homeowners, who are ‗house-rich‘ but ‗cash-poor‘, whereby they could liquidate
a portion or the whole of their home equity over a period of time to create a regular
stream of income.
Reverse mortgages are in existence in countries such as the US, Canada and Australia.
With the strained solvency of the state sponsored system, the onus in these countries has
shifted to private savings for retirement. Shifting the onus to private savings needs some
structural changes in the financial sector. Households will need to insure against
uncertainty related to real estate prices, rates of appreciation of property, interest rate
movements, inflation, and adjust for increasing longevity. While increasing the level of
precautionary savings can do this, this tends to reduce the level of disposable income
available for consumption, and therefore reduces net welfare. Shifting the risk to the
insurance agents, who are better equipped to handle this risk, is a more efficient solution.
A pareto optimal solution is for the individual to hold a part of his wealth as annuities to
aid in asset decumulation. (Yaari, 1965).
This paper attempts to investigate the effect of introducing a reverse mortgage into a
lifetime asset and liability scenario, measuring the impact of the reverse mortgage
product in terms of the net welfare increase for the individual. The introduction of a
product like the reverse annuity mortgage helps in smoothing consumption over the
lifetime of the individual by allowing her to tap into long term assets and by helping to
separate the consumption and the investment components of housing.
The rest of the paper is organised as follows: Section II gives a brief overview of
literature, section III discusses the Model development, Section IV defines the data
collection and Section V gives an overview of the results. This is followed by the
limitations and conclusions of the current study.
3
Literature Survey
The idea of smoothing consumption over the lifetime has had more than its share of
research, both theoretical and empirical. There is however, no single accurate
explanation, although the life cycle models of Modigliani and Brumberg (1954), Ando
and Modigliani (1963) form the basis, with or without a bequest motive. Phelps (1962),
Samuelson (1969), Merton (1969), Hakansson (1970), have also put forward their
theories of portfolio decision making in an infinite horizon with certainty. On the
contrary, Yaari (1965), Levhari and Mirman (1977), and others studied the impact of
uncertain lifetime on planned consumption with actuarially fair insurance.
After imposing the simplifying assumption that interest rate is zero, Modigliani and
Brumberg (1980) arrived at the individual consumption function. Because the retiree
finances his consumption by running down the assets accumulated over the working life,
wealth holdings follow a humpshaped age profile with a peak at the end of the earning
span as stressed in Modigliani (1986).
Deaton (1992) asserts that the introduction of a positive interest rate does not
significantly change the main features of the life cycle model. In the beginning of the life
cycle consumption path shifts downwards, whereas at the end it shifts upwards. Income
path is more realistically described by a hump-shaped age profile. During the early years
of working career income is typically low, but increases along with ageing. Eventually
earnings decline at the end of the working career. According to Deaton (1992), in this
case consumption smoothing implies that a young worker might in fact want to borrow
rather than to save during the early years of his earning span.
Zeldes (1989b) proved that once the benchmark life cycle model is substituted by a
model with random labor income and constant relative risk aversion utility function
(isoelastic utility function) the low dissaving of the elderly accords with rational
optimizing behaviour. In a model like this, the agent is allowed to insure oneself against
future contingencies, such as health shocks, with saving motivated by precaution. With
respect to the behaviour of an economic agent, the implications of the CRRA utility are in
general more plausible than those of quadratic utility.
4
Among the proposed causes for the slow dissaving of the elderly is uncertainty about the
date of death. Hurd (1989, 1990) has shown using a life cycle model with uncertain
lifetime that slow decumulation of wealth is associated with a rather smooth consumption
path. According to his interpretation, modest current consumption and postponed
decumulation are a means for a highly risk averse consumer to hedge against the
possibility of having insufficient resources if he should live unexpectedly long.
Davies (1981), on the other hand, claims that uncertainty over the length of life depresses
consumption by a fraction that increases with age. The reduction in consumption then
leads to lack of dissaving in old age.
Kotlikoff and Spivak (1981), Kotlikoff et al. (1987) and Abel (1985) also find that with
longevity being certain, or in an actuarially fair market, people tend to consume much
more in their retirement phase, following the consumption smoothing stipulated by
finance theory, and it is the risk aversion of individuals and uncertainty of their life-span
that causes them to leave unintended bequests.
The presence of housing in the portfolio, however, skews the patterns of investment and
consumption predicted by the above models. The investment in housing is much larger
than should be the weightage given to a single asset, given that housing is a fairly risky
proposition. Economic and finance theory looks at housing as a deviation from the
standard models of asset allocation. For most homeowners, the house is the single largest
asset in their portfolio. Housing differs from stocks and bonds and other investment
assets because a house combines both investment and consumption. The household's
ownership of residential real estate determines the level of its consumption of housing
services, and the level of real estate ownership that is optimal as consumption good is
different from the level of real estate from an investment viewpoint.
With developed rental markets, a household can separate the consumption and the
investment aspects. However, rental housing is by no means a perfect substitute for
owner-occupied housing, because of various agency and transaction costs over the
lifetime of home ownership. Housing can thus be reduced to a make-or-buy decision.
Buying a house insulates the consumption of housing services from variation in the rental
price of housing, but housing is an illiquid asset and subject to liquidity and interest rate
risks. Homeowners‘ wealth is exposed to house price risk whereas renters‘ wealth is not.
5
In terms of taxation, owner occupied housing services are not taxed and mortgages
interest payments are deductible from the income tax base. The two most important
housing tax advantages are that the service income provided by owner-occupied housing
(generally referred to as imputed rents) is untaxed and mortgage interest payments are
deductible from taxable income. (Gervais, 2002).
However, owned housing as a hedge against rent risk. Institutional and borrowing
constraints frequently prevent young households with low levels of cash in hand from
purchasing a house that matches their lifetime consumption need. Senior homeowners, in
the meantime, are often forced to hold an equity position in their houses that lasts longer
than their expected length of occupancy. This mismatch between life cycle housing
consumption need and housing investment position is worsened by the presence of lumpy
housing adjustment costs.
Jan K. Brueckner (1997) analyzes the interaction between the consumption demand and
the investment demand for housing in a mean-variance portfolio model. Brueckner
considers a general covariance matrix and mean vector of returns and a general utility
function, and derives analytical results. Flavin and Yamashita (2002) derive the portfolio
choice of level of housing stock and why this impacts the bonds to stock ratio for
households at different age and wealth levels, and solve for housing stock in a mean
variance efficiency framework, using numerical methods and simulation.
Scope of the Study
The scope of this paper is to introduce reverse mortgage into such a scenario, specific to
the Indian context. Introduction of reverse mortgage into this framework changes the
types of risks that are faced by the participants. In an instance where the home owner
does not want to shift homes for various reasons, this model will derive a solution using
simulation of house prices, appreciation, rate of increase in income, etc to identify
whether buying a house initially and then buying a reverse mortgage is a more optimal
solution than buying the house without the reverse mortgage, or not buying a house at all.
A reverse mortgage is described as an annuity payment to the homeowner for the length
of time that she remains in the house. No repayments have to be made on the loan as long
as the homeowner is alive and still living in the house affected by the reverse mortgage,
6
or till the spouse of the homeowner is alive and still living in the said property. This is
better than conventional types of mortgage where the homeowner has to either move out
or sell her property, or has to undertake to make regular annuity payments on a loan.
The valuation of the reverse mortgage and the monthly or annual receipts from it depend
on the age of the homeowner, the value of the home, the amount of loans (if any) on the
property, and even the location of the property. The homeowners can choose to receive
the reverse mortgage funds as a lump sum, as monthly income (for up to life), as a line of
credit, or any combination.
This paper analyses housing as a consumption and investment decision, in a life-cycle
model without a bequest motive. In a typical portfolio of equity and bonds, housing is
introduced. Equity has a yield that mimics the stock market behaviour in India, bonds
have a deterministic yield, and housing has both consumption and investment aspects
modelled based on the imputed rent and the home appreciation rates respectively. The
owner of the house makes payments on a Fixed rate mortgage in the initial part of his life,
and towards retirement, draws down from the home equity as a reverse mortgage. The
paper uses simulation methods to solve a life-cycle model under alternative assumptions
about the proportions of equity, debt and housing in the portfolio of the individual.
The study aims to explore the potential wealth effects of Reverse Mortgage by
considering the extent to which housing equity could change the consumption of the
elderly, if housing equity were converted to a more liquid form. The mechanism for home
equity conversion that is considered is a reverse annuity mortgage. The primary goal is to
understand the potential effect on the current income of the elderly through conversion of
illiquid housing wealth into an income stream.
Taking a median income level and a housing asset accumulation and decumulation
pattern that is in keeping with the practices in India, the increase in wealth due to
availability of a reverse mortgage instrument is contrasted with the life cycle wealth in
the absence of such an instrument. In particular, the increase in current income of elderly
families if they were to obtain reverse mortgages, and thus convert housing equity into
cash that could be used for other consumption, is studied. By this, it is hoped that a strong
case can be made for the introduction of reverse mortgage products in India. Using both
7
mortgage and reverse mortgage in the lifetime housing asset plan, homeowners may be
able to stabilize levels of expenditure through the lifetime.
Data
The paper uses primary data from Housing development Finance Corporation, (HDFC), a
leading provider of housing finance in India. Primary Data from HDFC covers data on
the house mortgagee profile, age, income levels, the amount of housing equity the owner
plans to invest in, the tenure of the loan, Adjustable or Fixed rate mortgage, and the
amount of down payment. The Income-house equity distribution for various states in
India is derived from the empirical distribution and values are generated from the same
for the simulation.
The yield of the equity market is calculated from the S&P CNX NIFTY, which is an
index of the National Stock Exchange, India. The returns on bonds are taken as the
returns on the Treasury Bills (180 days) published rate of return from the Reserve Bank
of India, the Central Bank in India. World Bank, UN ESCAP data are available on
population, ageing, pension coverage, etc. The National Account Statistics and NCAER-
SEBI Study on the Indian Investor are used extensively to calculate the distribution of
wealth/income across different categories of income and to cross-verify the same.
We have assumed growth rate of 5% as property appreciation rates. The actual long-term
property appreciation rates in India are not very well documented, so we make the
assumption that property prices are indexed to inflation prevalent at the same time period.
Also, growth rates in property are different across different cities as well as within
different zones in the same city. In India, there isn‘t a refined concept of zoning of areas,
so within each cluster of area, the property appreciation rates may be vastly different.
The analysis is done during the 1995 to 2005 period. Average pay scales denominated by
the Pay commissions, a government body that decides pay scales for various classes of
employees are used in arriving at average monthly pay, housing loan affordability and
repayment, and pension accruals. Average monthly income levels for the elderly are
computed based on Census data and cost of living indices. A reverse mortgage plan is
then incorporated into the wealth model to identify the increase in income due to easy
liquidation of housing wealth.
8
Monte Carlo simulation is used to conduct the above analyses. Sensitivity to various
parameters such as home loan interest rates, Public Provident Fund interest rates,
differing levels of income and home valuations are also studied.
Model Development
Scenarios
The model compares the lifetime utility of a scenario including housing and reverse
mortgage compared to scenario in which there is an option of only housing and no
reverse mortgage.
Comparing three scenarios
Scenario – H + RM
Scenario 2 – H + No RM
Scenario 3 – H + No RM + moving to a smaller home and utilising the difference in
equity.
Scenario 4 – H + No RM + selling the house and utilising the proceeds as an annuity.
Lifetime Asset Allocation
The individual receives exogenous labour income of non-stochastic, certain nature from the
time of start of his work life till his retirement at the fixed age of sixty. (Sixty is t he age of
mandatory retirement in India for those in government and most private services). Since
Income after primary employment is not certain, these incomes are not considered. They
can, however be easily modelled in the existing framework.
The individual invests such income that he receives from labour in both financial and
non-financial assets (real assets including housing). The investor‘s portfolio then consists
of housing, investments in stocks and bonds and in other non-appreciating assets.
Investors dynamically decide on house size, tenure, financial portfolio, and other goods
consumption. The lifecycle simulation begins between the ages of 20 and 30 (uniform
distribution of cases between 20 and 30), which is the year of start of the work-life of the
individual. Each period of analysis is one year, and mortality of the individual is
uncertain.
9
The model is characterised by a representative single-person household that plans its optimal
consumption of housing and non-housing services, and the optimal timing of first home
ownership, over its lifetime, given perfect foresight about its future income.
The simulation model applied here is a modified housing tenure choice model in which the
decisions by a household about when to buy a first home, the size of the purchase and
subsequent upgrades, consumption and saving pre and post retirement are made jointly in a
life cycle framework.
A life cycle approach is appropriate because the desire to pay off a home provides an
incentive to save. The life cycle simulations provide certain insights. Policies to improve
housing affordability can simultaneously alter the timing of the household‘s first home
purchase, the size of the purchase and therefore house prices and the level of private saving.
The simulations suggest that if households bring forward their first home purchase – that is,
they buy at a younger age – the size of the purchase will be smaller due to tighter borrowing
constraints implied by their lower income and lower accumulated savings.
Model Variables
Time periods
Assume individual starts work life at time period t, works for duration of N years and has
life expectancy of E post retirement.
Total life of the individual = t+ N+E,
Where ‗t‘ is the age at which the person enters the workforce, N is the work-life of the
individual and E is the life-expectancy at age 60 for the individual.
Income
Income of the individual is It at any point in time. It is a function of appreciation in
income and inflation. Nominal It is a function of Real It and inflation. For the purpose of
simulation, nominal It will be used. The growth rate of personal disposable income as
calculated from the National Account Statistics of India (NAS, 2000-2005, various years)
is used as the compound rate of growth of personal income. Since the NAS gives details
on the rate of growth of nominal income, inflation is not considered separately.
Preconditions
There is no asset or liability for the individual at Time t, that is, there is no hereditary
wealth, outstanding loans etc. Any such positive or negative wealth can be considered as
exogenous to the model and model can be adjusted accordingly to account for this by
10
considering ‗t‘ as a variable. Positive wealth can be treated equivalent to a scenario in
which the individual has started working earlier to ‗t‘ and negative wealth can be
considered as a scenario in which the person effectively starts working at ‗t‘. ‗t‘ can thus
be chosen at a point where net assets is equal to zero.
Work life of the individual
Consider that the individual has a productive work life of ‗N‘ years with W0 defining the
start of work life. Wn will thus be the nth year of work life. The individual retires at WN,
and WN can also be considered start of retirement , so WN = R0. Re is the eth of retirement
and RE is the life expectancy of the individual at age (t+ N), which is the year of
retirement.
Age t +N can be considered to be 60 for the Indian scenario. Sensitivity analysis can be
performed on this variable also.
Mortality
Since we are interested in the welfare effects of a reverse mortgage, which is applicable
only over the age of sixty, we do not model mortality before the age of sixty. The life
expectancy at age sixty is factored into the model. Mortality data at age 60 is obtained
from the Life Insurance Corporation of India.
Housing variables
Mortgage
Let M0 be year of start of mortgage. At the time of start of mortgage, a downpayment of
Hd is made on the property. So, at the time of initiation of mortgage, ALL the savings of
the individual (savings being equivalent to investments in stocks and bonds) is made as
down payment of the house. The savings in other assets, such as gold and fixed
mandatory saving vehicles, or those in which there is no option of early withdrawal is not
taken into consideration. (these include, but are not limited to life insurance policies,
National Savings Certificates, Public Provident Funds, investments in Gold and other
durable assets).
Let Mt denote the tth year of mortgage. The mortgage terminates at MT where T denotes
the tenure of the mortgage. At MT, the housing equity of H is transferred to the owner of
the property. The model limits itself to cases in which pre-payment doesn‘t occur. Since
11
prepayment is normally a function of transfers of wealth and the model doesn‘t consider
any transfers of wealth – this is a reasonable assumption to make. Labour income is the
only source of income considered and the consumption function is assumed to be
constant throughout the lifetime of the individual.
The rate of interest on the mortgage is fixed throughout the life of the mortgage (Fixed
rate mortgage) and the payout on EMI is, therefore, fixed.
Tenure
The simulation is based on the distributional parameters of the empirical distribution
generated from the data on the age-profile of homeowners. Tenure choices for the home
loans also follow the empirical distribution. Tenure is also subject to two conditions, one
that the mortgagee fully owns the house before retirement at sixty, and second, the EMI
payments for that particular tenure choice do not exceed the income-to-outflow ratio of
35% fixed by industry norms. Prepayment of the housing loan is not allowed, and tenure
choices are carried over till the end of the term.
Downpayment
An investor enters into a housing loan for purchase of a house when he has the resources
for down payment on the house size he wants to purchase. Following industry norms in
India, this has been fixed at 15% of the housing equity. The amount of down payment
will decide the size of the house that can be purchased. There is literature available to
suggest that young couples decide on their home equity levels based on the amount of
down payment that they can make. In practise, there is an option to defer the purchase of
home equity in case the resources are not available for the desired house size. The
simulation however does not have the option of deferring the purchase of the home.
If households accelerate their home purchase, they may be forced to buy a smaller house
size relative to their needs and preferences, because at a younger age they have less
accumulated saving to meet the down payment and also have lower income from which to
service the mortgage. Mortgage lenders set the maximum loan size according to the capacity
to meet monthly repayments, which is called the Installment-to-income ratio that is assessed
mainly on the basis of the flow of personal disposable income.
Housing Consumption and Imputed Rent
Both implicit housing consumption (when the owner-occupied house is fully paid off)
and explicit interest payments are counted as owner-occupied housing consumption.
12
Imputed rent is considered a saving for the owner-occupied house and the imputed rent is
calculated from the lifetime annuitisation of the value of home equity (the maximum
value of home equity is equal to how much the individual would pay for consuming
housing services of a similar nature over his lifetime). This disregards holding a housing
asset for investment purposes, but this is not a restrictive assumption to make.
A younger homeowner can enjoy greater tax benefits over the lifetime, - in particular, the
tax-free status of the imputed rental income from owner-occupancy. This implies higher
lifetime resources for consumption, and per period consumption is greater than for an
older homeowner.
Housing services may be obtained either by purchasing a house or by renting housing. In
the initial years (during working life) households purchase rental housing till they are
able to afford their own housing. It is assumed that housing purchases take place at the
end of a period, and with each period being one year; the household must wait at least
until the end of the first period to buy a house.
Appreciation of property
It is assumed that houses do not depreciate, but the value of land + housing appreciates at
the rate of ‗house appreciation‘ Ha = (1+ha). Again, there are very rare cases in which the
total housing unit depreciates in value (as in a ‗housing bubble‘) but the simulation works
for negative values of appreciation as well. In this study, the value of housing
appreciation is assumed to be 5%, which is just the rate of inflation Wholesale Price
Index) in India, and sensitivity analysis is performed for other rates of appreciation.
Disposal of the Housing Asset
Post retirement, the household has three options
1. To remain in the same house and finance consumption out of retirement
income/wealth
2. To reverse mortgage the house and to finance consumption out of the reverse
mortgage annuity payments and the wealth
3. To sell the house
In case of option (3), two options exist- move to another house – typically with EMI
payment on the house, or with a down payment on purchase, or to rent an
accommodation. The rent paid on the second accommodation and the EMI option are
13
equivalent in nature and substitutable, provided there are no other transaction costs and
the discount rate on the EMI is the same as the interest rate for borrowing and lending.
This is however a restrictive assumption to make, since transaction costs involved in
selling a house are about 7% and transaction costs in purchase of a new accommodation
(including registration etc) are about 15% of the value of the new accommodation.
In the presence of perfect capital markets, no transaction costs, no liquidity premiums,
perfect information etc, however an asset is held, it will achieve its fair market value on
sale. Consumption also would be perfectly smoothed over the lifetime and there would
not be any welfare loss across the three options outlined above. Since transaction costs
are one of the most important determinants of welfare loss, it is important to identify and
incorporate transaction costs as part of the simulation. Data on transaction costs (stamp
duty etc) is given in the appendix.
Consumption function
Literature shows consumption function to be a function of initial income, interest rates
inflation rates variance in income economic shocks If we assume that labour income
being the primary source of income is fairly stable over the lifetime of the individual, and
labour income also adjusts dynamically to the inflation rate and to the rate of growth of
GDP (implicit interest rates) then consumption function is reduced to being only a
function of initial income class
It = Ct + Sat
It is income, Ct is consumption and Sat is savings at time ‗t‘.
Savings
The normal avenues for savings are financial assets such as stocks and bonds and durable
physical assets such as housing and Gold. The NSSO data gives by MPCE class
investment in stocks and bonds. Also, housing loan companies cap the mortgage amount
based on the income levels. It is reasonable to assume individuals will consume the
maximum amount of housing they are able to afford at any point in time.
Sat = Bt + St + Mt + Otherst
BT = bti*Bt where b is (1+ rate of interest on bonds- after tax)
ST likewise denotes the cumulative value of wealth in stocks at year T
14
The maximum amount of housing depends on
a. the amount of down payment
b. maximum mortgage amount available to them through the Instalment-to-income
ratio.
Pension
After retirement, individuals receive a pension, given by a fraction P%r > 0 of average
labor earnings in a particular year. In India, only about 11% of the population receives
any pension from government or quasi-government bodies. The scenario with pension is
used to determine whether the individuals receiving any kind of pension also benefit from
reverse mortgage. %r is the average endowment during an individuals working life from
his labor earnings. P is a replacement rate and %r corresponds to indexed average annual
earnings. Typically pensions are financed by tax on the individual, which is the personal
tax rate of 35% in India.
EMI= Max (Ih, Instalment-to-income ratio), where Ih is the EMI calculated as a function
given by
EMI = (Mh, Tenure, rate of mortgage)
Total Housing equity is given by HT = HD+ Principal component of Mh
Where t =T , Mh=0
HD is to be minimum 15% of total house equity, and at the point of origination of
mortgage, HD is greater than total savings in stocks and bonds. (Down payment is
financed totally by own funds, and no borrowing is allowed)
Retirement
At R0, total wealth of the individual is given by the wealth in stock, bond, other assets
and the appreciated value of housing wealth.
If the individual decides to sell the property and move to a new one, transaction costs of
7% are applicable on the sale and 15% on the acquisition of a new property. Sensitivity to
various levels of realistic transaction costs is analysed. In case of sale of property and
renting of a new one, imputed rent is taken as a consumption post the retirement phase.
The imputed rent is slightly difficult to calculate: the methodology used is to assume that
the individual consumes the same rent as earlier, but because there is a lesser time period
15
to mortality, the value of housing purchased will be equal to the annuitized value of rent
for the remaining lifetime of the individual. In case where the individual decides to rent
the property, the same amount of imputed rent will be treated as consumption
expenditure.
Reverse Mortgage
Using actuarial calculations, it is possible to calculate the value of the house for a
specified tenure of reverse mortgage.
Assume the ownership of the house reverts totally to the bank on the death of the owner,
but during the lifetime of the owner, he enjoys occupancy of the house. Suppose the
current value of the house is H, then the value of the house will appreciate by ‗g‘ every
year. The owner, who is now of a particular age ‗a‘ will die at time T with a probability
tpa that can be calculated using the mortality tables. Mortality tables used as reference in
this are from LIC A(1996-1998) Mortality table for annuitants. The value of the loan to
the bank at current date is obtained by discounting the value of this house with the
mortgage rate ‗m‘. The lump sum amount that the borrower can borrow from the bank is
given by the actuarially fair valuation:
t a
T
1 g
L [ H * *t pa ]
t a 1 m
T here is the maximum age till which a person can live, taken to be 120, based on the LIC
A (96-98) Mortality table for Annuitants. L is the lump sum the bank will pay to the
owner for immediate title to the house.
The lender can then determine the actuarially fair maximum present value of an annuity
(LS) that a borrower can purchase, and break this down into a stream of monthly
payments to the borrower, calculated using the following, where ‗r‘ is the risk free rate.
L
P t a
T
p
1t ar
t a
We have assumed an interest rate of 8% in computing the monthly payments. We also
report the findings for 8%, 10%, and 12% interest rates. The median interest rate on a
16
similar annuitant policy provided by the Life Insurance Corporation of India, the leading
insurer in the country, works out to about 8.5%.
Results
The simulation of life-cycle asset accumulation and decumulation patterns was performed
using Matlab, version 6.0 and @Risk to generate the distributions points from the
distribution parameters.
Given below are some details about the primary data
Loan Term
Income Rate of Interest Property Value
(Months)
Mean 31901.59 9.75 1240546.67 205
Median 19122.00 10.5 870000.00 216
Std. Deviation 56732.09 1.55 1675739.97 94.39
Minimum 2028.00 7.25 33000.00
Maximum 3052555.00 13.25 55000000.
Loan Term
Percentiles Income Rate of Interest Property value
(Months)
10 9269.5000 7.50 376320.00 84
20 11427.0000 8.00 484624.00 120
30 13702.0000 8.50 599572.00 156
40 16120.0000 9.00 715000.00 180
50 19122.0000 10.50 870000.00 216
60 23290.0000 10.75 1084978.00 240
70 28800.0000 11.00 1342564.20 240
80 37380.0000 11.25 1694763.20 267
90 56845.0000 11.25 2337000.00 310
The Median income is about 20,000 per month, or about 2,40,000 INR per annum.
Income figures are reported as Net income. The EMI is calculated based on the property
value less the value of down payment (which averages to 15% for the sample) and the
rate of Interest.
The median value of the property is 8,70,000 INR and the property value is restricted by
the amount of EMI that can be paid per month on the availed loan amount. Most of the
loanees appear to limit their down payment to the mandatory 15%. This indicates, in all
probability, that the amount of housing equity investors want to invest in is limited by the
17
Amount they can afford as down payment. This has also been documented in literature,
where it is said that investors choose the size of the housing asset based on assets
available as down payment.
The Cross-tab of income and housing is given below.
Empirical Probability Distribution of Income vs. Property Values
Income /
Property 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 33 37 41 45 49
1 0.05 0.39 0.42 0.10 0.02 0.01
2 0.02 0.12 0.28 0.25 0.16 0.08 0.05 0.02 0.01 0.01
3 0.03 0.08 0.11 0.15 0.16 0.14 0.12 0.08 0.04 0.03 0.02 0.01 0.01 0.01
4 0.02 0.04 0.07 0.08 0.09 0.12 0.14 0.11 0.11 0.06 0.05 0.04 0.03 0.01 0.01 0.01 0.01
5 0.01 0.01 0.04 0.06 0.07 0.12 0.13 0.13 0.10 0.07 0.08 0.05 0.04 0.03 0.03 0.01 0.01
6 0.03 0.06 0.04 0.06 0.06 0.10 0.06 0.10 0.08 0.08 0.06 0.07 0.04 0.06 0.05 0.02 0.01
7 0.02 0.02 0.04 0.07 0.04 0.06 0.07 0.11 0.09 0.05 0.08 0.05 0.07 0.09 0.08 0.03 0.03 0.02
8 0.03 0.04 0.03 0.08 0.05 0.06 0.10 0.06 0.06 0.07 0.06 0.04 0.06 0.09 0.06 0.04 0.02 0.03
9 0.01 0.03 0.03 0.04 0.03 0.06 0.04 0.08 0.09 0.08 0.03 0.05 0.11 0.15 0.09 0.05 0.03
10 0.04 0.06 0.06 0.04 0.04 0.06 0.03 0.09 0.06 0.06 0.09 0.12 0.09 0.04 0.04 0.04
11 0.02 0.05 0.00 0.07 0.07 0.04 0.07 0.13 0.07 0.04 0.05 0.13 0.16 0.07 0.02
12 0.02 0.06 0.06 0.08 0.08 0.12 0.04 0.06 0.02 0.02 0.06 0.14 0.08 0.06 0.06 0.02
13 0.03 0.05 0.05 0.08 0.05 0.03 0.03 0.03 0.03 0.05 0.29 0.13 0.05 0.08 0.03
14 0.04 0.07 0.04 0.07 0.04 0.15 0.04 0.07 0.15 0.15 0.07 0.04 0.07
15 0.07 0.04 0.04 0.11 0.04 0.07 0.11 0.04 0.04 0.07 0.04 0.11 0.04 0.14 0.04 0.04
17 0.03 0.03 0.05 0.03 0.03 0.05 0.05 0.03 0.05 0.13 0.13 0.10 0.18 0.03 0.03 0.13
20 0.03 0.09 0.03 0.06 0.06 0.03 0.06 0.06 0.06 0.21 0.06 0.06 0.06 0.12
23 0.07 0.11 0.00 0.04 0.14 0.07 0.11 0.21 0.07 0.04 0.11 0.04
26 0.04 0.04 0.04 0.04 0.04 0.08 0.04 0.21 0.08 0.08 0.25 0.04
29 0.06 0.06 0.13 0.13 0.13 0.06 0.19 0.19 0.06
Souce: Generated from primary data
Note: Income value in Lakhs per annum and Property value in Lakhs per annum
The joint distribution of housing vs income shows that housing is linearly increasing with
income, but there is a lot of variation within each group. The matrix is fairly diagonal in
nature. The Bayesian probability here is conditional upon the income level of the
individual. Where there has been a joint ownership of the house, the income of both the
spouses has been taken into consideration.
18
Income vs Housing Wealth
0.50
0.40
0.30
Probability
0.20
0.10 Property
Value
0.00 11
1.00
5.00
9.00
1
13.00
18.50
Income 0.00-0.10 0.10-0.20
0.20-0.30 0.30-0.40
0.40-0.50
Assume that individuals‘ preferences over consumption streams are represented by
N (C t )(1 )
E t i
t 0 1
where is a subjective discount factor and is the relative risk aversion coefficient.
The term inside braces indicates the constant relative risk aversion function.
Given the life-cycle hypothesis about consumption smoothing of the individual, any
deviation from projected consumption can be taken as a loss in welfare. Representing this,
Welfare loss = ( Cit Ca )2
t
Where Ci is the estimated consumption and Ca is the actual consumption at time ‗t‘.
There are two parts to the welfare measure: One is whether the timing of house purchase
makes any difference, the second is whether the decision to go for a reverse mortgage
makes a significant difference.
Timing of house purchase
The timing of house purchase makes a difference in terms of utility of the individual.
Individuals who invest in house equity earlier rather than later in life have a significant
difference in terms of wealth.
19
Age of House purchase Smoothed Consumption through
the lifetime (per annum)
25 101,345
30 94,580
35 89,078
40 85,142
45 81,287
Source: Calculated from primary data using Monte Carlo Simulations
Note: Marginal tax rate as applicable in India in 2006-07, Income of Rs.2,00,000 (median
income) and house property of Rs 10,00,000. (Median property value.) and term of 240
months (median value)
Income levels Marginal Tax Rate
< 110,000 None
110,000<Income 150,000 10% of amount greater than 110,000
150,000<Income 250,000 20% of amount greater than Rs. 150,000 + Rs. 4,000
Income 250,000 30% of amount greater than Rs. 250,000 + Rs. 24,000
The reason why investment in a house is an attractive option is because there are tax
breaks (both principal and interest payable on housing loan can be offset against tax
liability) and the ownership of the home acts as a hedge against house rent appreciation.
The descriptive statistics of the rates on financial assets in India is as given below
Asset Mean Rate of return
Stock Index (S&P CNX NIFTY) 13.9%
Treasury Bill (Risk Free Rate) 4.5%
Mortgage Loan 9.75%
Inflation 5%
House Price Appreciation 10%
The simulation of lifetime wealth suggests that investing in a housing asset is first order
dominant over not investing in a housing asset. During the accumulation phase, the tax
offset provided by housing increases net wealth to the individual due to investing in a
housing asset.
20
Decumulation Phase
The decumulation phase involves a choice of
a. Selling the housing asset at the current market price of the house and renting or
buying a new house
b. Using a reverse mortgage
c. Not tapping into house equity at all
Selling a house at current market price indicates that in the scenario of both renting a
house or buying a new one, the net present value of the proceeds from house equity are
reduced by the net present value paid out as rent or as Equated monthly installment on the
new house. So both of these scenarios are equivalent.
Using a reverse mortgage is a more efficient solution when the net transaction costs from
reverse mortgage are lesser than transaction costs on the option of selling the house.
There are also other psychological reasons that are involved in the decision to see the
house. Also, in case the individual decides to sell the existing property, she exposes
herself to rent appreciation risk, and to interest rate risk. If she remains in the same
property and goes in for a reverse mortgage, the individual is no longer exposed to rent
appreciation risk, but interest risk still exists. In any case, if rent moves in tandem with
property prices, then the individual stands to gain from house price appreciation rather
than lose if she chooses a reverse mortgage over selling and renting a house.
Either way, the welfare loss is least on opting for a reverse mortgage over not opting for a
reverse mortgage. Comparing the scenario of selling vs. reverse mortgaging the house,
reverse mortgage leads to greater utility depending on the parameters considered.
Payment amounts in a RM depend on the following
Higher Home Equity (value of home less any outstanding indebtedness) increases
the amount a homeowner can take out of the home equity
Older Households experience greater income increases due to the shorter life
expectancy. Also, joint life expectancy is much lesser than individual life
expectancy
Lower interest rates increase the amount per payment because cumulative interest
to be repaid is lower
21
Higher expected property appreciation will increase the payment amount because
the value of the remaining home equity is higher
Lower inflation rates, because these work to keep the nominal interest rate lower
The larger the equity and greater the appreciation expected, the lower the risk that the
loan principal and accrued interest will exceed the market value of the property when the
loan becomes due. Higher interest rates mean more interest will accrue, increasing the
risk that the loan obligation will exceed collateral value. The longer the owner lives, the
greater the loan obligation whether the borrower chooses the annuity or lump sum with
accrued interest option.
The principal limit factors for reverse mortgage at different ages is as given below.
Interest Rate Age of Borrower at Start of RM
65 75 85
8.00% 0.537 0.626 0.728
10.00% 0.361 0.492 0.638
12.00% 0.261 0.389 0.560
Risk Factors in a Reverse Mortgage
Given the nature of the product, the long term impact of interest rates and appreciation
in real-estate prices, there are a lot of risk factors that are associated with a product like
reverse mortgage.
Lenders still face the risks of longevity, interest rate changes, and future property values
(Szymanoski, 1994). The uncertainty of borrower longevity and tenure makes it difficult to
estimate the timing of repayment and, therefore, the return the loan will yield. The interest costs
on outstanding balances follow a variable rate pattern, with certain limits on the rate at which the
interest costs can change within a year, and over the life of the product.
The more serious risk that the reverse mortgage provider bears concerns the growth in
the principal limit factor over time. There is "crossover risk", which is realized when the
principal limit factor exceeds around 90 percent of the house value. From this point on,
the lender becomes unable to recover the full amount owed.
The risk faced by borrowers in a reverse mortgage is also substantial. Borrowers do not
have the same capacity to diversify that lenders have. For a homeowner, there is a trade-
22
off between using housing wealth as a hedge against lump sum cash outflows such as
those for medical expenses or long-term care, vs. using housing wealth to append to
existing sources of income. This trade-off is important in the decision to go for reverse
mortgage. Elderly homeowners put off the decision of reverse mortgage since there is no
external motive, which forces them to a decision.
The risk faced by borrowers on a reverse mortgage includes the ability to predict
definitively their cash requirements at various times. In elderly homeowners this
becomes more important since they are more likely to incur costs, which are of a non-
recurring and fairly unpredictable nature. Reverse mortgage also in some cases includes
a lump-sum amount, but in this case, it is more expensive than taking a loan with the
home as a collateral.
Welfare from Reverse Mortgage
Comparison of welfare provided by reverse mortgage over the last ten years of life with
the option of no Reverse mortgage (selling and renting or buying another house) shows
that a reverse mortgage is a better solution than the other two scenarios considered.
Comparison of Welfare Functions
20
15
Welfare
10
5
0
1
3
5
7
9
11
13
15
17
19
21
Number of cases
Average Welfare in the last ten years of life with RM
Average Welfare in last ten years of life without RM
We perform 20,000 simulations and finally, we calculate the mean and standard error
and confidence intervals of over all the iterations
23
Across income groups, the welfare gain remains similar, but the difference between the
welfare with and without RM is different. Lower income quartiles seem to benefit more
from the reverse mortgage than higher income quartiles, as has also been shown in
various other studies by Venti and Wise etc. (1991)
Average Welfare for the Highest Income Quartile
20
15
Welfare
10
5
0
1 2 3 4 5 6 7 8 9
No of Cases
Average Welfare for the last ten years of life with RM
Average Welfare for the last ten years of life without RM
Average Welfare for Lowest Quartile Income Levels
20
15
Welfare
10
5
0
1 2 3 4 5 6 7
Number of Cases
Average Welfare in last ten years of life with RM
Average Welfare in last ten years of life with RM
24
Welfare Gains by introducing Reverse Mortgage to smooth consumption fluctuation
Percentile = 10 =5 =1
100% 0.068 0.039 0.023
70% 0.062 0.034 0.020
50% 0.059 0.028 0.018
30% 0.043 0.019 0.013
10% 0.020 0.012 0.004
Note: Return on Stocks is 10%, bonds is 5%, home appreciation at 10%, personal discount rate of 10%.
Looking at the data, one can conclude that unlocking housing equity does have a
substantial impact on the income levels of individuals. However, there are some
limitations to this study, which are discussed below.
The true cost (ex post) of obtaining funds from a reverse mortgage will depend on the
timing of the borrower‘s death—if the borrower lives well past his life expectancy then a
reverse mortgage will have low costs but if he dies sooner than expected then a reverse
mortgage will be expensive.
Limitations of the Model
The simulation model has a number of strong assumptions that necessarily qualify the
results. The most important ones are
a. Income is certain and of non-stochastic nature. There are no income shocks which
means the household unit smoothes consumption over the lifetime based on the
certain income levels. However, even if there is an income shock that is of a
certain nature, the model can account for it, since households are assumed to
smooth consumption based on all available information. So if the shock were
known in advance the household would take that into account while smoothing
consumption over the lifetime.
b. It is a partial equilibrium model that implies the household has no consumption
shocks as well. Consumption in each period is known in advance, and if the
predicted consumption pattern does not occur, welfare loss is bound to be lower
than calculated welfare loss. The results will still hold in terms of which
alternative provides the highest welfare.
25
c. There is a specific distribution of house equity to income levels existing based on
the current norms followed in India. Since this joint distribution of house equity
with income is derived from empirical data, any change in the financing norms of
housing companies, and or sentiments of people will change this distribution.
d. Also, currently, the tenure choice for housing loans is also taken from empirical
data. This limits the tenure to a discrete distribution, mostly on multiples of five
years. The same limitation previously discussed applies to this in terms of
generating data from an empirically observed distribution.
e. There is no uncertainty in house appreciation. Sudden variations in property
prices are not accounted for, and a nominal growth rate of property is assumed
which is expected to last throughout the lifetime of the household. Housing
bubbles, as well as the seasonality in house prices that is documented in western
countries is not taken into account here
f. The imputed rent is taken to be the rent which is payable for a house of similar
size, and this imputed rent is taken from National Sample Survey data. Imputed
rent is different for different locations, and the empirical data on housing equity
obtained may be for different locations, with distributions of rent being different
from that captured in the National Sample Survey.
Despite these limitations, the life cycle framework is the appropriate way to analyze
housing and saving decisions because these are fundamentally joint decisions made in a
life cycle context. The simulation model indicates that households will respond to ways
to decumulate assets from housing
The development of the housing mortgage market combined with the development of
reverse mortgage as an instrument for elderly homeowners throws open a wide variety of
opportunities. People who traditionally regard real estate as a safe investment are likely to
invest in housing equity so that they have a regular rent stream during the years when the
mortgage exists and a regular income drawn them from home equity during the period of
the reverse mortgage.
In summary, reverse mortgages appear to have a lot to offer a limited number of elderly
households. It offers a means of accessing hard earned equity and raising the quality of
26
lifestyle in later years. The reverse mortgage industry in India, when it starts, can benefit
from the experience of overseas housing markets, especially in the USA, and tailor a product
that will suit the elderly household, the financier, and society at large. With rapidly growing
superannuation and insurance funds, this appears to be an ideal opportunity to tap into the
security of a residential mortgage whilst increasing the overall standard of living.
Conclusion
An insight that emerges from this model is that households are better off investing in
a house due to the tax benefits of housing during the asset accumulation phase. Even
if we consider the house appreciation rate (investment return rate on housing as an
asset) to be lower than that on stocks and nearly equivalent to the inflation adjusted
risk free rate of return, we find that because of the implicit advantages of housing
(tax benefits, plus zero outflow towards rental value after the purchase of the house)
investing in the house is a first-order dominant result. This result mimics those that
have been obtained in studies on housing earlier.
Of the three different scenarios analyzed in terms of decumulation, the sale and purchase
of a new property leads to the greatest welfare loss for the investor. This is because, the
transaction costs that are involved are very high, at about 22-25% of the total value of the
proceeds. The reverse mortgage fares best of all in most cases, with the option of selling
the house and taking one more on rent being the second best. Of course, these will be
sensitive to assumptions made about the portfolio composition and the house size that is
acquired after retirement.
We find that for most income levels except the highest income class, reverse mortgage is
the instrument of choice. While this study models only the monetary factors of reverse
mortgage, the psychological factors involved in moving out from a house are not
considered.
27
Appendix
Figure 1 Composition of Household Wealth
Household Asset Composition
50
40
Percentage
30
20
10
0
Livestock
Building
Receivables
Durable
Financial
Machinery
Land
Assets
HH
1981 1991 2002
Source : Households Assets and Liabilities in India, NSSO, NSS 59th Round, Jan-Dec 2003.
Table I Type of Assets held by the Elderly
Male Female
Assets 1994-1995 1986-1987 1994-1995 1986-1987
Rural Urban Rural Urban Rural Urban Rural Urban
Having And 569 581 450 398 177 185 168 114
Financial Managing
assets All 695 702 581 525 391 376 482 371
Having And 651 605 639 539 206 206 228 167
Property Managing
All 804 742 819 700 456 420 633 481
Source : Aged In India, NSSO, NSS 52nd Round, 1995-96
28
Table 2 Household Sector Physical and Financial Savings
(At current prices)
Old Series (Base: 1993-94)
Year Household Sector Savings (in Crore) Household Sector Savings (in percent of GDP)
Financial Physical Total (1+2) Financial Physical Total (3+4)
Savings (1) Savings (2) Savings (3) Savings (4)
1970-71 1371 3263 4634 3.0 7.1 10.1
1971-72 1555 3664 5219 3.2 7.5 10.7
1972-73 2128 3496 5624 3.9 6.5 10.4
1973-74 3612 4373 7985 5.5 6.7 12.2
1974-75 2374 5706 8080 3.1 7.4 10.4
1975-76 3918 5825 9743 4.7 7.0 11.7
1976-77 4852 6997 11849 5.4 7.8 13.2
1977-78 5853 8501 14354 5.8 8.4 14.1
1978-79 6658 10357 17015 6.0 9.4 15.4
1979-80 6081 10609 16690 5.0 8.8 13.8
1980-81 8610 11258 19868 6.0 7.8 13.8
1981-82 9614 11611 21225 5.7 6.9 12.6
1982-83 12739 10477 23216 6.8 5.6 12.3
1983-84 13294 14871 28165 6.1 6.8 12.8
1984-85 17879 17188 35067 7.3 7.0 14.3
1985-86 18538 21257 39795 6.7 7.6 14.3
1986-87 23336 21736 45072 7.5 7.0 14.5
1987-88 26820 32337 59157 7.6 9.1 16.7
1988-89 27183 43474 70657 6.4 10.3 16.8
1989-90 37998 48957 86955 7.8 10.1 17.9
1990-91 49640 60257 109897 8.7 10.6 19.3
1991-92 62101 48635 110736 9.5 7.4 17.0
1992-93 65367 65706 131073 8.7 8.8 17.5
1993-94 94738 63572 158310 11.0 7.4 18.4
1994-95 120733 78625 199358 11.9 7.8 19.7
1995-96 105719 110421 216140 8.9 9.3 18.2
1996-97 141661 91591 233252 10.4 6.7 17.1
1997-98 146777 121660 268437 9.6 8.0 17.6
1998-99 180346 146456 326802 10.4 8.4 18.8
1999-00 205743 198658 404401 10.6 10.3 20.9
2000-01 216774 235494 452268 10.4 11.3 21.6
2001-02 253964 259146 513110 11.2 11.4 22.6
2002-03P 254439 320242 574681 10.3 13.0 23.3
2003-04QE 314261 357431 671692 11.4 13.0 24.3
New Series (Base : 1999-2000)
Year Household Sector Savings (in Crore) Household Sector Savings (in percent of
GDP)
Financial Physical Total (1+2) Financial Physical Total
Savings (1) Savings Savings (3) Savings (4) (3+4)
(2)
1999-00 206602 210124 416726 10.5 10.7 21.3
2000-01 215219 231098 446317 10.2 11.0 21.2
2001-02 247476 255198 502674 10.8 11.2 22.0
2002-03 253256 312152 565408 10.3 12.7 23.1
2003-04P 316444 332190 648634 11.5 12.0 23.5
2004-05QE 320777 366302 687079 10.3 11.7 22.0
P: Provisional. QE: Quick Estimates.
Source: RBI.
29
Table 3 Changes in Financial assets/Liabilities of the Household Sector
(At current prices)
Year Financial Curren Bank Non- Life Provi- Claims Shares Units Net
Assets cy Deposits Bank Insur- dent and on and of UTI Trade
# Depo- ance Pension Govern- Deben- Debt
sits ## Fund* Fund ment+ tures++
(1-9) (1) (2) (3) (4) (5) (6) (7) (8) (9)
1970-71 2110 355 754 67 207 490 105 68 14 50
1971-72 2319 404 1024 104 251 474 -2 20 12 32
1972-73 2982 637 1214 108 307 523 80 27 19 67
1973-74 3578 769 1511 45 356 603 87 -16 24 199
1974-75 3371 18 1654 92 344 787 72 62 -3 345
1975-76 5067 342 2120 130 423 1224 899 41 16 -128
1976-77 6651 1140 3920 114 524 1172 19 -5 20 -253
1977-78 7154 703 3521 227 592 1316 325 201 34 235
1978-79 9483 1430 4626 232 683 1605 227 204 79 397
1979-80 10249 1332 4659 477 773 1748 531 253 41 435
1980-81 12118 1625 5550 378 915 2122 712 412 31 373
1981-82 13621 965 5194 894 1037 2480 1784 510 114 643
1982-83 16097 2026 6661 870 1235 2865 1243 646 122 429
1983-84 18790 2776 7978 1019 1376 3052 1976 555 222 -164
1984-85 23549 2938 9859 960 1556 3759 3107 762 567 41
1985-86 25562 2220 10603 1423 1779 4188 3413 1394 586 -44
1986-87 31849 3090 14510 1512 2159 5055 3092 1768 943 -280
1987-88 36106 4815 14674 1326 2589 6509 3680 813 1196 504
1988-89 39958 4256 14747 1580 3423 7552 5478 1136 1427 359
1989-90 48233 7655 13987 1839 4415 9508 6758 2655 2179 -763
1990-91 58908 6251 18777 1286 5599 11155 7883 4972 3438 -453
1991-92 68045 8157 17848 2218 7003 12501 4845 6800 9087 -414
1992-93 80354 6562 29518 6035 7114 14814 3885 8212 5612 -1398
1993-94 109618 13367 36236 11654 9548 18323 6908 10067 4705 -1190
1994-95 145501 15916 55835 11547 11370 21414 13186 13473 3908 -1148
1995-96 124337 16525 39941 13198 13894 22343 9588 8839 262 -252
1996-97 158519 13643 50902 25980 16121 30390 11783 6631 3776 -708
1997-98 171740 12780 74099 6733 19410 32267 22162 4464 595 -770
1998-99 207103 21822 79433 7670 23428 46408 28220 5105 1887 -6870
1999-00 236214 20845 82892 3844 28644 53907 28985 16308 1811 -1023
30
2000-01 248394 15632 94703 6911 33861 47882 39007 11148 -934 183
2001-02 296581 28156 112936 7912 41237 46609 51938 9834 -1857 -183
2002-03 322583 28632 123462 8788 52009 48441 56087 7122 -1618 -341
2003-04P 380090 42675 141967 3803 52240 51655 87372 9078 -8586 -114
2004-05P 435706 36977 158259 3370 69572 56354 106420 8113 -3146 -213
2005-06$ 588656 51954 274641 4567 83340 58615 86755 29452 -444 -222
P: Provisional.
$: Preliminary Estimates.
# Includes deposits with co-operative non-credit societies.
## Due to changes in coverage of non-banking deposits, data prior to 1997-98 are not strictly comparable with those of
1997-98 and onwards.
* Includes state/central government and postal insurance fund.
+ Includes compulsory deposits.
++ Includes investment in shares and debentures of credit/non-credit societies, public sector bonds and investment in
mutual funds (other than UTI).
Source: RBI.
31
Table 4 Changes in financial Asset/Liabilities of the Household Sector in Percent
(At current prices)
Year Change Curr- Bank Non- Life Provi- Claims on Shares Units Net
in ency Deposits Bank Insur- dent and Govern- and of UTI Trade
Financial Depo- ance Pension ment Deben- Debt
Assets sits Fund Fund tures
(1-9) (1) (2) (3) (4) (5) (6) (7) (8) (9)
1970-71 100.00 16.82 35.73 3.18 9.81 23.22 4.98 3.22 0.66 2.37
1971-72 100.00 17.42 44.16 4.48 10.82 20.44 -0.09 0.86 0.52 1.38
1972-73 100.00 21.36 40.71 3.62 10.30 17.54 2.68 0.91 0.64 2.25
1973-74 100.00 21.49 42.23 1.26 9.95 16.85 2.43 -0.45 0.67 5.56
1974-75 100.00 0.53 49.07 2.73 10.20 23.35 2.14 1.84 -0.09 10.23
1975-76 100.00 6.75 41.84 2.57 8.35 24.16 17.74 0.81 0.32 -2.53
1976-77 100.00 17.14 58.94 1.71 7.88 17.62 0.29 -0.08 0.30 -3.80
1977-78 100.00 9.83 49.22 3.17 8.28 18.40 4.54 2.81 0.48 3.28
1978-79 100.00 15.08 48.78 2.45 7.20 16.93 2.39 2.15 0.83 4.19
1979-80 100.00 13.00 45.46 4.65 7.54 17.06 5.18 2.47 0.40 4.24
1980-81 100.00 13.41 45.80 3.12 7.55 17.51 5.88 3.40 0.26 3.08
1981-82 100.00 7.08 38.13 6.56 7.61 18.21 13.10 3.74 0.84 4.72
1982-83 100.00 12.59 41.38 5.40 7.67 17.80 7.72 4.01 0.76 2.67
1983-84 100.00 14.77 42.46 5.42 7.32 16.24 10.52 2.95 1.18 -0.87
1984-85 100.00 12.48 41.87 4.08 6.61 15.96 13.19 3.24 2.41 0.17
1985-86 100.00 8.68 41.48 5.57 6.96 16.38 13.35 5.45 2.29 -0.17
1986-87 100.00 9.70 45.56 4.75 6.78 15.87 9.71 5.55 2.96 -0.88
1987-88 100.00 13.34 40.64 3.67 7.17 18.03 10.19 2.25 3.31 1.40
1988-89 100.00 10.65 36.91 3.95 8.57 18.90 13.71 2.84 3.57 0.90
1989-90 100.00 15.87 29.00 3.81 9.15 19.71 14.01 5.50 4.52 -1.58
1991-92 100.00 11.99 26.23 3.26 10.29 18.37 7.12 9.99 13.35 -0.61
1991-92 100.00 11.99 26.23 3.26 10.29 18.37 7.12 9.99 13.35 -0.61
1992-93 100.00 8.17 36.73 7.51 8.85 18.44 4.83 10.22 6.98 -1.74
1993-94 100.00 12.19 33.06 10.63 8.71 16.72 6.30 9.18 4.29 -1.09
1994-95 100.00 10.94 38.37 7.94 7.81 14.72 9.06 9.26 2.69 -0.79
1995-96 100.00 13.29 32.12 10.61 11.17 17.97 7.71 7.11 0.21 -0.20
1996-97 100.00 8.61 32.11 16.39 10.17 19.17 7.43 4.18 2.38 -0.45
1997-98 100.00 7.44 43.15 3.92 11.30 18.79 12.90 2.60 0.35 -0.45
1998-99 100.00 10.54 38.35 3.70 11.31 22.41 13.63 2.46 0.91 -3.32
1999-00 100.00 8.82 35.09 1.63 12.13 22.82 12.27 6.90 0.77 -0.43
32
2000-01 100.00 6.29 38.13 2.78 13.63 19.28 15.70 4.49 -0.38 0.07
2001-02 100.00 9.49 38.08 2.67 13.90 15.72 17.51 3.32 -0.63 -0.06
2002-03 100.00 8.88 38.27 2.72 150.00 15.02 17.39 2.21 -0.50 -0.11
2003-04P 100.00 11.23 37.35 1.00 13.74 13.59 22.99 2.39 -2.26 -0.03
2004-05P 100.00 8.49 36.32 0.77 15.97 12.93 24.42 1.86 -0.72 -0.05
2005-06$ 100.00 8.83 46.66 0.78 14.16 9.96 14.74 5.00 -0.08 -0.04
P: Provisional.
$: Preliminary Estimates.
# Includes deposits with co-operative non-credit societies.
## Due to changes in coverage of non-banking deposits, data prior to 1997-98 are not strictly comparable with those of
1997-98 and onwards.
* Includes state/central government and postal insurance fund.
+ Includes compulsory deposits.
++ Includes investment in shares and debentures of credit/non-credit societies, public sector bonds and investment in
mutual funds (other than UTI).
Source: RBI.
33
Table 5 Gross Financial Saving of the Household Sector in Percent
Item 2005- 2004- 2003- 2002- 2001- 2000- 1999- 1998- 1997- 1996- 1995- 1994- 1993-
06# 05P 04P 03P 02 01 2000 99 98 97 96 95 94
Financial Saving 100 100 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0
(Gross) (16.7) (14.0) (13.8) (13.1) (12.7) (11.9) (12.2 (11.9) (11.3) (11.6) (10.5) (14.4) (12.8)
)
a) Currency 8.8 8.5 11.2 8.9 9.7 6.3 8.8 10.5 7.4 8.6 13.3 10.9 12.2
(1.5) (1.2) (1.5) (1.2) (1.2) (0.7) (1.1) (1.3) (0.8) (1) (1.4) (1.6) (1.6)
b) Deposits 47.4 37.0 38.3 40.9 39.4 41.0 36.3 38.8 46.6 48.1 42.5 45.5 42.6
(7.9) (5.2) (5.3) (5.4) (5.0) (4.9) (4.4) (4.6) (5.3) (5.6) (4.5) (6.5) (5.4)
i) with banks 46.7 36.4 37.4 35.5 35.3 32.5 30.8 33.7 37.8 25.7 26.3 35.3 27.9
ii) with NBFC 0.8 0.8 1.0 2.7 2.6 2.9 1.7 3.8 3.9 16.4 10.6 7.9 10.6
iii) with co-op 0 0.0 0.0 2.8 3.6 5.6 4.3 4.6 5.3 6.4 5.8 3.0 5.2
banks/societies
iv) Net trade debt 0 0.0 0.0 -0.1 -2.1 0.1 -0.4 -3.3 -0.4 -0.4 -0.2 -0.8 -1.1
c) Shares and 4.9 1.1 0.1 1.7 2.7 4.1 7.7 3.4 2.9 6.6 7.3 11.9 13.5
debentures (0.8) (0.2) (0.0) (0.2) (0.3) (0.5) (0.9) (0.4) (0.3) (0.8) (0.8) (1.7) (1.7)
i) pvt. corporate 1.3 1.4 1.1 0.8 1.5 3.1 3.4 1.5 1.3 3.6 6.6 8.0 7.5
ii) co-op banks/ 0 0.0 0.0 0.0 0.1 0.0 0.0 0.1 0.1 0.1 0.1 0.1 0.1
societies
iii) units of UTI -0.1 -0.7 -2.3 -0.5 -0.6 -0.4 0.8 0.9 0.3 2.4 0.2 2.7 4.3
iv) bonds of PSUs 0.0 0.0 0.0 0.1 0.0 0.1 0.1 0.0 0.1 0.1 0.1 0.1 0.5
v) mutual funds 3.6 0.4 1.2 1.3 1.8 1.3 3.4 0.8 1.1 0.3 0.3 1.1 1.2
(other than UTI)
d) Claims on 14.7 24.4 23.0 17.4 17.9 15.7 12.3 13.6 12.9 7.4 7.7 9.1 6.3
Govt. (2.5) (3.4) (3.2) (2.3) (2.3) (1.9) (1.5) (1.6) (1.5) (0.9) (0.8) (1.3) (0.8)
i) investment in 2.4 4.9 7.5 2.5 5.8 1.7 0.9 0.7 1.6 0.4 0.4 0.1 0.4
Govt. Securities
ii) investment in 12.3 19.5 15.5 14.9 12.1 14.0 11.3 13.0 11.3 7.0 7.4 9.0 5.9
small savings etc
e) Insurance 14.2 16.0 13.7 16.1 14.2 13.6 12.1 11.3 11.3 10.2 11.2 7.8 8.7
funds (2.4) (2.2) (1.9) (2.1) (1.8) (1.6) (1.5) (1.3) (1.3) (1.2) (1.2) (1.1) (1.1)
i) Life ins. funds 13.5 15.1 13.0 15.5 13.5 12.9 11.2 10.6 10.6 9.5 10.4 7.2 8.0
ii) Postal ins. 0.2 0.3 0.3 0.3 0.3 0.2 0.3 0.3 0.3 0.3 0.3 0.2 0.2
iii) State ins. 0.5 0.6 0.5 0.4 0.4 0.5 0.6 0.5 0.4 0.4 0.5 0.5 0.5
f) Provident and 10.0 12.9 13.6 15.0 16.10 19.30 22.80 22.40 18.80 19.20 18.00 14.7 16.7
pension funds (1.9) (1.8) (1.9) (2.0) (2.0) (2.3) (2.8) (2.7) (2.1) (2.2) (1.9) (2.1) (2.1)
# Preliminary.
P: Provisional.
Note: Figures in brackets are percentages to GDP at current market prices.
34
Source: RBI
Figure 2 Trend of Inflation Rate in India
Inflation Rate - India
14
12
Inflation Rate
10
8
6
4
2
0
1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004
Year
Source: ADB Country Estimates, 2004
Figure 3 Bank Term Deposit Rates
Bank Term Deposit Rates
14
12
Interest Rates
10
8
6
4
2
0
1970
1972
1974
1976
1978
1980
1982
1984
1986
1988
1990
1992
1994
1996
1998
2000
2002
2004
Source : RBI and various issues of Handbook Of Statistics on Indian Economy
35
Figure 4 MPCE Class-wise Average Expenditure Incurred for Acquiring New Residential
Unit During Last 5 Years for All Urban Areas in India
Average Expenditure per New Residential Unit
900
Average Expenditure (in '000s)
800
700
600
500
400
300
200
100
0
0-300 350-425 500-575 665-775 915-1120 1500-1925
MPCE Class
Source : indiastat.com
Figure 5 Mortality Rates by Age
Mortality Rates by Age
0.6
0.5
Mortality Rates
0.4
0.3
0.2
0.1
0
105
112
14
21
28
35
42
49
56
63
70
77
84
91
98
0
7
Age
Source: LIC A (96-98) Mortality Table for Annuitants
36
Table 6 Construction Activity in Indian Cities
Building Completion Certificates Issued During the Years
Million Plus City
1994 1995 1996 1997 1998 1999 2000 2001
Ahmedabad 636 181 519 370 244 211 107 187
Bangalore 3707 3735 4517 4549 4583 4615 4648 4616
Bhopal 2027 2453 1273 2252 2815 1405 2144 2121
Chennai 4333 5380 5986 5941 5760 5485 5573 5606
Coimbatore 397 1087 941 1120 1136 1153 1171 1154
Delhi 595 834 715 1015 1019 916 882 939
Greater Mumbai 2846 2886 1001 758 468 742 737 649
Hyderabad 4983 5160 5310 5497 5692 5995 6359 6015
Indore 607 1530 4223 5562 6208 3831 3083 4374
Jaipur 837 872 1592 1657 1725 1795 1869 1796
Kanpur 1418 1453 1966 2013 2060 2110 2159 2109
Kochi 2282 2301 2367 2391 2410 2434 2458 2435
Kolkata 2604 2680 4011 4128 4249 4372 4499 4373
Lucknow 1413 1500 2104 2232 2368 2513 2663 2515
Ludhiana 1029 717 850 1712 1805 1907 2013 1908
Madurai 952 962 645 1281 1299 1317 1335 1318
Nagpur 268 277 676 696 716 737 759 738
Patna 731 743 784 797 811 825 839 826
Pune 1092 1116 1047 1273 1201 1174 1216 1197
Surat 102 1574 1682 1796 1917 682 618 1072
Vadodara 1322 1135 864 1244 1613 2610 2655 2293
Varanasi 714 1243 1416 1448 1295 1081 1110 1162
Visakhapatnam 1233 1268 1297 1335 1374 1649 1644 1556
Total 36128 41087 45786 51067 52768 49559 50541 50959
Source : indiastat.com, 2001
37
References
1. Ando, A. and F. Modigliani (1963). "The 'Life Cycle' Hypothesis of saving: Aggregate
Implications and tests." American Economic Review 53(1): 55.
2. Brueckner, Jan K.(1997) "Consumption and Investment Motives and the Portfolio Choices of
Homeowners." Journal of Real Estate Finance and Economics, 75(2):159-80.
3. Deaton, Angus S., 1991, ―Savings and Liquidity Constraints,‖ Econometrica 59, 1221—1248.
4. Deaton, Angus S., 1992, ―Understanding Consumption,‖ Oxford University Press, Oxford.
5. Flavin, M., Yamashita, T. (2002),"Owner-Occupied Housing and the Composition of the
Household Portfolio", The American Economic Review, Vol. 92(1): 345-362.
6. Gervais, M. (2002). "Housing taxation and capital accumulation." Journal of Monetary Economics
49(7): 1461-1489.
7. Government of India (2002), India 2002: A Reference Annual, Publications Division, Ministry of
Information and Broadcasting, New Delhi.
8. Hakansson, N. (1969) ―Optimal Investment and Consumption Strategies under Risk, an Uncertain
Lifetime, and Insurance‖. International Economic Review, 10 (3): 443-466.
9. Hurd, Michael D. (1989), Mortality risk and bequests, Econometrica, 57(4),779-813.
10. Kotlikoff, L. and Spivak, A. (1981) ―The Family as an Incomplete Annuities Market‖ The Journal
of Political Economy, 89 (2): 372-391.
11. Kotlikoff, L. and Summer, L. (1981) ―The Role of Intergenerational Transfers in Aggregate
Capital Accumulation‖. The Journal of Political Economy, 89 (4): 706-732
12. Kutty, Nandinee K., (1998), ―The Scope for Poverty Alleviation among the Elderly Home-owners
in the Unites States through Reverse Mortgages‖, Urban Studies, Vol 35 (1), 113-129
13. Ma, Seungryui, and Cho, Deokho (2005), ―A Lifetime Housing Asset Plan using Mortgage and
Reverse Mortgage Finance‖, Presented at the 2005 World Risk and Insurance Conference, Utah.
14. Levhari, D., L. J. Mirman, et al. (1980). "Capital Accumulation Under Uncertainity" International
Economic Review 21(3): 661.
15. Merton, R. (1969) ―Lifetime Portfolio Selection under Uncertainty: the Continuous-Time Case‖.
The Review of Economics and Statistics, 51 (3): 247-257.
16. Ministry of Finance, Government of India, Economic Survey 2002-2003, New Delhi.
17. Ministry of Social Justice and Empowerment Government of India (2000), OASIS 2000: Old Age
Social and Income Security- Report of the Project OASIS, New Delhi.
18. Modigliani, F. and Brumberg, R. (1954) ―Utility Analysis and the Consumption Function: An
Interpretation of Cross-Section Data‖. Post Keynesian Economics. Edited by Kenneth K.
Kurihara, 388-436
19. National Sample Survey Organization. Socio-economic profile of aged persons. Sarvekshana,
(1991) XV, Nos. 1-2, Issue No. 49.
38
20. National Sample Survey Organization. July 1995-1996 Aged in India: A Socio-economic Profile.
NSS fifty Second Round, Report No:446.
21. Phelps, E. (1962) ―The Accumulation of Risky Capital: A Sequential Utility Analysis‖.
Econometrica, 30: 729-743.
22. Samuelson, P. (1969) ―Lifetime Portfolio Selection by Dynamic Stochastic Programming‖. he
Review of Economics and Statistics, 51 (3): 239-246
23. United Nations (1987). ―Global trends and prospects of the age structure of population: Different
paths to ageing‖, in Papers and Proceedings of the United Nations International Symposium on
Population Structure and Development, (New York, United Nations).
24. United Nations (2000). United Nations Technical Meeting on Population, Ageing, and Living
Arrangements of Older Persons: Critical Issues and Policy Responses, (New York, United
Nations).
25. United Nations (2001a). World Population Prospects: The 2000 Revision, Volume I:
Comprehensive Tables, (United Nations publication, Sales No. E.01.XIII.8).
26. United Nations (2001b). World Urbanization Prospects: The 1999 Revision, (United Nations
publication, Sales No. E.01.XIII.11).
27. United Nations (2002). World Population Ageing 1950-2050, (United Nations publication, Sales
No. E.02.XIII.3).
28. United Nations (2004), Transitions in World Population, 2004, Population Reference Bureau,
United Nations
29. United States Social Security Administration (1999).
<http://www.ssa.gov/statistics/ssptw/1999/english>.
30. World Bank (2001), India: The Challenge of Old Age Income Security, Finance and Private Sector
Development, South Asia Region, World Bank, 5 April
31. Yaari, M. E. (1965). "Uncertain Lifetime, Life Insurance, and the Theory of the Consumer."
Review of Economic Studies 32(90): 137.
32. Yao, Rui, and Harold H. Zhang (2005), Optimal consumption and portfolio choices with risky
housing and borrowing constraint, Review of Financial Studies, 18(1), 197-239.
39
