University of Strathclyde, UK
1
Housing Policies and Urban Land Values in China: Lessons fromLatin America and Singapore
Roger J Sandilands
University of Strathclyde, UK
Conferencia de la IU, Buenos Aires, 15 de Mayo de 2012
ABSTRACT
This paper addresses the challenges facing China in accelerating the pace of rural-urban
migration, with special attention to the impact on income distribution and housing costs that
result from pressure on urban land values. It explains the push and pull influences on
migration and why its continuation is justified by the large gap between rural and urban
incomes and the relatively higher income elasticity of demand for urban-based goods and
services. The provision of affordable housing is an integral part of this structural shift
programme. The paper thus considers the most appropriate ways in which housing finance
can be mobilised, but supplemented by the capture of urban land values on behalf of the
whole community so both the quality and affordability of housing can be increased. Positive
and negative lessons for China are offered from the different urbanisation and fiscal policy
experiences of Latin America (especially Colombia) and Singapore.
1. Introduction
Despite unprecedented economic growth in recent decades, accompanied by huge
demographic shifts to her cities and accompanying residential construction, China
undoubtedly will need continuing rapid expansion of its urban housing stock for many years
to come. Thirty years after switching to a more market-oriented economy, China attained by
2008 a per capita income estimated at US$2,940 or US$6,020 in terms of purchasing power
parity (World Bank, 2010). For countries at this level of development the typical proportion of
the population living and working in urban areas is around 55% (Henderson 2009:6); but for
China the proportion is much lower at only 44.9% in 2010 (United Nations, 2011). The
proportion was less than 20% in 1980 at the start of China’s reform period (having risen little
during the volatile Maoist era from a mere 13% in 1950). Much has been achieved since
then, but while rapid economic growth has clearly gone hand-in-hand with rapid
urbanisation, until recently this urbanisation has actually been at a significantly slower pace
(3-4% per year) than the 5-6% rates typically achieved by other developing countries during
their periods of rapid economic growth (Choi 2000; World Bank 2002).
Choi (2000), Henderson (2009), World Bank (2009, 2012), Gao (2010) and many others
highlight that this relatively modest record is partly due to legal restrictions on rural-urban
migration associated with the hukou system that has been motivated by a desire to avoid the
crowded shanty-town slums and associated social unrest prevalent in many cities in Latin
America, Africa and other parts of Asia.
2
One consequence of these restrictions, however, is
that the rural-urban income gap in China is one of the widest in the developing world. Choi’s
(2000) estimate for the World Bank put the gap at about 2.5:1 between 1978-1998, and it
appears that the gap has been widening since then. According to the World Bank (2010:66)
it had risen from around 2.6 in 1999 to around 3.3 by 2006, and World Bank (2012: 301)
1
Paper to be presented at the biennial conference of the International Union for Land Value Taxation,
Buenos Aires, May 15, 2012 on “Housing, Land and Social Inclusion.” I have drawn liberally on a
paper I originally prepared for the Asian Development Bank (ADB), July 2011. The views expressed
do not necessarily reflect the views and policies of the ADB.
2
Nevertheless, according to the Population Reference Bureau (2010), 38% of China’s urban
population in 2005 lived in what are described as slums. (The comparable figure for Colombia was
18%, and zero for Singapore.) 2
estimates that the overall Gini index of inequality has increased from around .30 in 1980 to
.49 in 2007, and that income and wealth disparities have also been increasing within the
expanding urban areas. The present paper suggests that one of the major causes of these
trends is the unequal capture of high and rising urban land rents.
World Bank (2012:405) has forecast approximately a three-fold increase in real land prices
in China between 2005 and 2030, or a real annual average rate of land price inflation of
around 6 per cent. The rate of increase will undoubtedly be much higher for urban land
alone. Unfortunately the World Bank’s analysis of the land market in China is vitiated by
remarks such as that “land along with investment has fuelled China’s extraordinary levels of
infrastructure development, urban development and industrial growth and has contributed to
maintaining social stability” (ibid: 136). First, this statement glosses over the widespread
instability and violent protests connected with the corrupt requisitioning of rural land in China.
Second, land does not produce wealth. Rather it is a passive recipient of wealth produced by
the active factors of labour and capital. Only when they are applied is wealth produced. Be a
country ever so rich in natural resources, without labour, capital, and knowledge, nothing is
produced. If land is fixed, production can only increase through more labour, capital or
technology.
Classical economists such as Henry George (1897, Bk. II, Ch. XIV) insightfully distinguished
value from production and value from obligation. Neo-classical economists ignore this
distinction and take all incomes as equal measures of productive contributions. Thus is the
private appropriation of monopoly rent justified. Worse, by conflating land with capital, as
they do, there is no rent, only profits from “enterprise”. In this way the neoclassical
economist avoids the ethical issue altogether. Naturally, this theory neither explains progress
nor why progress marches hand-in-hand with poverty.
However, to reduce the average rural-urban income disparity, a necessary precondition is
greater labour mobility from low-paying to higher-paying activities. To this end, Henderson
(2009), for example, would like to see the hukou system abolished, but accepts that there
must be a transition period so that migration is at a pace that would not add to the strains
already experienced by the most attractive cities. He suggests that a first step could be to
relax restrictions at the provincial level before relaxing them nationally. However, the more
that decent housing and jobs for potential migrants can be provided, the faster that hukou
restrictions can be relaxed without being disorderly.
This paper seeks (i) to explain the general economic forces that lie behind these structural
shifts, and (ii) to indicate the positive and negative lessons for the pace and pattern of
house-building and urban design that China can gain from the experience of Colombia and
Singapore at similar stages of their development, and (iii) to highlight the ethical and
economic implications of the privatisation of the tremendous land value increments that
accompany rapid growth.
2. The main economic forces that drive urbanisation
The shift in the structure and centre of gravity of countries as they develop economically is
conventionally explained by grouping the influences into “push” and “pull” forces. However,
the distinction is not clear-cut because these forces interact on each other. One can speak of
the distinct forces of supply and demand but one sector’s supply is ultimately its demand for
the goods and services provided by other sectors in exchange.
Nevertheless, consider first the side of demand. In the early stages of development low
incomes force people to spend the bulk of their subsistence incomes on food, basic clothing
and basic shelter. Agriculture is then the dominant sector and its land-intensive nature forces
people to live in scattered communities that satisfy their basic needs at the level of the 3
household or village, with relatively little specialisation and market-based exchange. The
lack of specialisation keeps labour productivity low.
3
As incomes grow, more is spent on food but less as a proportion of income. The proportion
rises for other goods and services. This is the near-universal “Engel’s Law” that can be
expressed as a tendency for the income-elasticity of demand for food and raw materials to
be significantly less than unity while for other things it is significantly higher (with the
weighted sum of the elasticities adding up to unity). As supply responds to demand, so
labour and capital are pulled toward the production of manufactured goods and related
services such as transport, education and health. Village settlements expand if they can
provide these more efficiently than on the farms; and as incomes and patterns of demand
alter in favour of non-agricultural products so villages become towns and towns become
cities that exchange their products for those of the farms and mines.
Next consider the supply side. Advances in technical knowledge, experience, skill, capital
accumulation and organisational and institutional change increase the direct and indirect
productivity of labour. Thanks to astonishing improvements linked with the Green Revolution
over the last 50 years, actual and potential productivity gains have been more powerful in
world agriculture than in most other economic sectors. The most notable gains are
associated with new high-yielding “miracle” seed varieties of rice, wheat and other crops that
can increase output per acre and per person six-fold or more as compared with traditional
varieties. Powerful gains also come from increased mechanisation and the complementary
use of chemical fertilizers, pesticides, irrigation and drainage.
Why, then, are the national (as opposed to best-practice) statistics on agricultural
productivity increases relatively disappointing in many developing countries, and what are
the implications for overall economic growth? A superficial answer is that there has been
under-investment in this sector. A deeper answer seeks the reasons for this “under-
investment” and notes that “markets clear at the short end”. That is, if productivity and
potential supply increase faster than the increase in effective market demand it follows that
prices will fall until the increase in supply matches the slow increase in demand. The point is
that although, by definition, productivity increases in agriculture reduce costs and that this
usually reduces prices – especially in highly competitive agriculture – the low elasticity of
demand for foodstuffs (even among poor people) means that the trend increase in demand
for agricultural output is typically only a little greater than the growth of population – perhaps
3-4% a year in most developing countries or somewhat more in countries like China where
overall growth of incomes is much higher.
The conclusion is that it is a mistake to identify revolutionary increases in productivity with an
increase in production, or sales, or farm incomes. When productivity increases outstrip
production and sales, it is because the lower prices of cost-reducing innovators have driven
higher-cost non-innovators out of the market, forcing them either to retire into subsistence
farming or to migrate to towns and cities in search of an alternative livelihood. This is just as
true if the innovations are in the form of (i) capital-intensive and labour-saving machinery
such as tractors and combined harvesters or (ii) the greater use of labour-intensive inputs
such as new yield-increasing seed varieties, fertilizers, irrigation and drainage. The latter
certainly enable more to be planted and harvested on the farms where they are used, and
these farms may well employ more labour as a result. But it is a fallacy of composition to
3
Adam Smith (1776) famously considered that increased specialisation or division of labour is the
most powerful way in which wealth increases, but that its extent is limited by the size of the market.
For this reason his magnum opus was primarily an attack on prevailing mercantilist policies that
restricted markets by setting the interests of producers above those of consumers. To this end, Smith
called for enhanced competition and mobility in both product and factor markets. The implications for
policy on rural-urban migration should be obvious but remain controversial. 4
assume that what is true of innovating farms is also true of the whole sector. Productivity
increases on the more commercial, modernising farms (usually the larger ones with greater
profits and access to credit) capture a larger and larger share of a relatively slowly
increasing market demand. This again implies that the smaller, non-innovating farmers lose
market share and are driven out. The net increase in overall farm output and measured
average productivity thus is less – often much less – than the increase in production and
productivity on the innovating farms where these are both increasing rapidly.
This explains the “push” or labour-displacing effects on migration arising from the supply
side of agriculture interacting with inelastic demand. An implication is that these effects can
be exacerbated if government policy subsidises agricultural investment rather than relying on
the natural response of investors to actual market demand; and this is true even if the
subsidies focus only on labour-intensive inputs.
However, there are simultaneous “pull” effects of increased agricultural productivity on urban
employment opportunities. Lower costs that reduce food and raw material prices benefit
consumers, especially those in urban work who do not rely on agriculture for their money
incomes. Urban-sector employers can even reduce the real wages they need pay to attract
workers and this enables them to employ more workers than otherwise.
4
Higher incomes
increase demand for agricultural products, even as it declines as a proportion of incomes. If
the resulting increases in farm incomes lead to higher investment in that sector, this may
further increase productivity ahead of the slowly increasing demand, thus further displacing
labour because of the relatively lower labour requirements per unit of output.
The question then is whether this dynamic “pull” effect from rising agricultural productivity on
urban employment dominates its “push” or labour-displacing effect. The fact that in many
countries, not least in China, the rural-urban wage gap has not significantly narrowed is
prima facie evidence that push has dominated pull, and that the pool of very low-productivity
rural workers remains so high that there is little upward pressure on rural wages. This in turn
depresses the wages of the marginal urban workers (mostly the recent immigrants) even as
the increased concentration of labour in more densely populated cities generates the
agglomeration economies that increase overall and average urban incomes.
If overall demand elasticity for urban goods and services is greater than unity, this means
that as urban productivity rises, the lower real costs and prices increase total expenditure on
these products. In contrast to agriculture, this impacts positively on urban employment
despite reduced labour requirements per unit of output. The income elasticity of demand for
urban services of all kinds (transport, distribution, finance, education, health, law,
administration and entertainments) is often somewhat higher than for urban industrial goods
while the productivity-increasing potential in these activities is generally lower than in both
agriculture and manufacturing. This combination explains why the share of services in both
GDP and overall employment increases as the economy grows. Similar comments could be
made about the construction sector, but here the experience is mixed, for reasons examined
in detail below.
5
Given the expectation of strongly positive employment-generating effects from the
combination of productivity growth plus high demand elasticities in urban sectors, what are
we to conclude if in practice the cities are failing to provide enough well-paying jobs for the
4
In reality most countries experience modest or high rates of general inflation, so that increased
productivity seldom leads to an actual fall in money prices. Rather, prices rise less than otherwise and
this allows urban employers to raise money wages by less than otherwise. This improvement in the
urban sector’s terms of trade with agriculture means that real wages fall from the urban employers’
perspective while still rising in terms of purchasing power from the employees’ perspective.
5
Connected with the possibility that demand is artificially repressed in this sector. 5
large and growing number of surplus workers who are either disguisedly unemployed in
agriculture or severely underemployed in the low-income urban informal sector?6
It is that in
agriculture the push effects are dominating its pull effects, and that the cities are failing to
exploit fully the opportunities that this underemployment represents for a better utilisation of
all available resources.
Insofar as these conditions still characterise China’s cities despite decades of strong growth,
this is the justification for an even greater pace of urban development there than has so far
been achieved. We may now show the positive and negative lessons may be gained from
policies pursued in Colombia and Singapore in their attempts to deal with similar problems of
underemployment, poverty and inequality.
3. The Colombian experience
For many years Colombia’s economic growth has been greatly below its potential when
considering how its abundant natural, physical and manpower resources have been
chronically underutilised. Population growth had doubled in only 20 years – from 11 million in
1953 to 22 million in 1973 – due to high rural birth rates coupled with a sharp decline in
infant mortality with the spread of vaccination programmes. Thereafter, population growth
began to decline, from 3.3% a year to 1.5% today, so the next doubling of population, to 44
million in 2005, took rather longer (32 years).
Population growth was accompanied by significant shifts to the cities. The urban population
was 33% of the total in 1950 and this had risen to 45% by 1960, a similar percentage to that
of China today. However, this figure was inflated by a violent civil war in the 1950s because
it was concentrated in rural areas which led to unusually strong outmigration. In the absence
of adequate conventional housing in central locations, a large proportion of these migrants
settled on the peripheries of the bigger cities, notably around the capital, Bogota, in
unlicensed squatter settlements or ‘barrios de invasión’, erecting flimsy structures and
illegally tapping electricity from nearby overhead cables.
7
Many of these settlements would
later be regularised by the authorities and provided with paved roads and vital services such
as water and sewage. Being distant from the major employment centres, workers had to
suffer long commuting times on overcrowded fume-filled buses.
What was clearly needed was a more adequate supply of shelter for actual and potential
immigrants into the cities. Its absence dilutes the ability of industry and commerce to prosper
by exploiting urban agglomeration economies. These economies are also restrained if there
is an inadequate supply of other structures such as factories, offices, schools and shops,
plus related infrastructure such as the road and utilities network.
All of these construction-sector projects have two things in common: they are long-lasting
investments and they are very costly in relation to current household incomes in the case of
housing; to current business incomes in the case of factories, offices and shops; and to
current state revenues in the case of public infrastructure. In the case of private sector
6
Disguised unemployment is the well-known term for workers who share a limited amount of work
and incomes but effectively produce nothing because the remaining workers could easily make up for
their work if they leave. In this underlying sense their marginal product is zero, while each worker’s
shared “institutional” income tends to the average product of the family or social group. Most studies
calculate the size of this surplus labour in a static sense by observing how many workers’ underlying
marginal product is zero under current conditions. This understates its potential dynamic extent as
and when workers are found alternative higher-paying jobs that increase demand for agricultural
products which thereby both motivates and provides the funding for labour-saving innovations.
7
The population of Bogota was approximately 1.7 million at the 1964 census and 2.9 million in 1973.
By the time of the 2009 census, it was nearly 8.6 million with a further million in nearby municipalities. 6
investments, this almost always necessitates outside funding over a long period of time
during which the buildings are providing their current services. Young families near the
beginning of their working lives can usually only buy a house or apartment costing two or
three times their annual income after supplementing their own limited savings with a
mortgage repayable over a long period. Similarly for new businesses. So these structures
are usually only affordable or desirable purchases if the monthly debt servicing costs are at
or below monthly rents for comparable properties.
This was the background to a national plan launched by President Misael Pastrana in 1972
as a bold “Plan of Four Strategies”. Its main author was Lauchlin Currie, a Canadian-born
New Deal economist who became President Franklin D Roosevelt’s personal White House
adviser for economic affairs from 1939-45. Currie first visited Colombia in 1949 as head of a
World Bank mission and stayed on to advise successive Colombian governments until his
death in 1993 (Sandilands 1990). He wrote the main background paper, Taming the
Megalopolis (Currie 1976), for the inaugural 1976 UN Habitat conference in Vancouver, and
drafted Recommendation D.3 of the Vancouver Action Plan on “Recapturing Plus Value” (in
Spanish, “plusvalia”): “The unearned increment resulting from the rise in land values
resulting from change in use of land, from public investment or decision or due to the general
growth of community”. As Currie (1976: 142-43) expressed the problem:
It is a striking example of our economic illiteracy that we have more or less quietly
acquiesced in the private appropriation of socially created gains, letting fortunate owners
and their heirs levy tribute or claim a share of the national income to which they have
contributed nothing… Generally, the case for capture of all or a large portion of the pure
monopoly gain of rising urban land has been impaired by failure to distinguish between
land and capital in general, between land and building, and between the rise reflecting
inflation and that traceable to pure scarcity.
Currie’s 1972 plan in Colombia was mainly designed to accelerate development via
measures that would greatly boost urban job opportunities. In brief, the four interrelated
strategies were (i) institutional innovations to boost urban housing via a new housing finance
system together with a new form of urban design; (ii) measures to boost non-traditional
exports by ensuring that the exchange rate would no longer be chronically over-valued; (iii)
increased agricultural productivity, to be accomplished partly through greater investment
expected to follow from higher demand resulting from faster growth of national income
together with the consolidation of rural farms as and when poorer farm families emigrated to
better-paying non-agricultural work; and (iv) improved income distribution related to the first
three strategies but intended to be complemented by more progressive taxation, notably by
the capture of rising urban land values via a “valorisation tax”.
The first priority (and the first of the four strategies) was to ensure that existing and future
urban families could enjoy far superior housing conditions than currently prevailed for the
majority. Hitherto, there had been three main types of housing and housing finance:
(i) Unlicensed sprawling settlements with overcrowded, unsanitary
shelters built with very poor quality materials. Finance for these very low-cost shacks
was generally out of the personal savings of the occupant families together with their
own work. They were usually on low-cost land whose value was low because of its
relatively remote location and/or because it lacked basic services. Sometimes these
‘barrios de invasión’ or squatter settlements were on valuable central-city sites being
held vacant either for speculative reasons or intended for buildings of a higher
standard that could use the land more intensively (because normally higher-rise
buildings of a kind that could not be built by unskilled amateurs).
(ii) Heavily subsidised popular housing for the very poor with funds provided by the
state-owned Instituto de Credito Territorial (ICT). These housing units were superior
to those typical of barrios de invasión but still generally of low quality, overcrowded, 7
and, because of the ICT’s financial constraints, also on low-value land that was not
necessarily the best location from a good urban planning perspective.
(iii) Subsidised higher quality housing for the middle and higher income groups financed
and built by the Banco Central Hipotecario (BCH) – or Central Mortgage Bank – that
held a virtual monopoly on this section of the mortgage finance industry.
Subsidies place a heavy direct or indirect strain on the national budget. In the case of both
the ICT and BCH, finance was raised via tax-exempt bearer bonds that favoured higher-rate
taxpayers, at the expense of government tax revenues. During the 1960s the average rate of
inflation was 11%, but fluctuated between 5% and 25%. Meanwhile, the nominal interest on
ICT and BCH bonds was fixed at 8.5% up to 1968 and 11.5% thereafter. Thus the real rate
fluctuated widely, was very low on average, and often negative. This deterred savings
except briefly between 1968 and 1971 when nominal rates were raised and inflation fell.
8
The government responded to the deficiency of saving by forcing public institutions such as
pension funds and other financial institutions to buy the unattractive BCH and ICT bonds.
In addition to these forced savings, the government directly subsidised the ICT out of the
central budget. These were necessitated by large losses on ICT operations leading to
chronic de-capitalisation. This was because the ICT could not charge a high enough interest
rate to its low-income borrowers to cover costs. Even with low or even negative real interest
rates delinquency on its loans was very high. (For example, the International Labour Office
(1970) reported that in 1969, 53% of the ICT’s loan portfolio was in arrears.)
This compared with a negligible delinquency rate at the BCH which, with its monopoly and a
focus on high-income customers, could usually cover its costs. However, for this it needed to
charge 16% on its mortgages, and restricted the typical mortgage term to 10 years because
of the risks associated with a high and very variable rate of inflation.
9
These conditions
significantly repressed demand for its mortgages, but this was in line with the relatively
sluggish growth of its available funds, and the BCH resisted change in its monopoly status
that insulated it from competitive pressures to grow.
The combined result of this housing system and related pattern of urbanisation was that the
growth of both commercial and social housing was every year falling far short of the growth
of new family formations in the towns and cities, hence continuously adding to an already
substantial housing deficit.
10
As for social housing (known in Colombia as ‘vivienda popular’
or ‘vivienda de interes social’) the self-built and ICT homes were either slums from the start
or, in the case of the ICT, were often to be the slums of the future because of their poor
quality, small size, poor location, inadequate amenities, and social segregation. Poor quality
of materials and low density also implied relatively high maintenance costs compared with
middle-income housing.
11
The only way to reduce the country’s growing housing deficit was to introduce changes that
would ensure an adequate volume of long-term loans on affordable terms such that the
growth of decent (non-slum) urban dwellings each year would exceed the growth of new
8
An important lesson was evidence of a very high interest elasticity of supply of savings. This made
possible a temporary boost to building in the late 1960s and also after the 1972 reforms.
9
This meant that nominal rates elsewhere in the market fluctuated more widely that the BCH rates,
creating an unstable feast-and-famine pattern to its inflow of funds.
10
According to the ICT, the urban housing deficit (defined as the difference between the number of
urban family units and the size of the urban housing stock, with consequent unhealthy overcrowding)
grew from 112,000 in 1951 to 254,000 in 1961 and 586,000 in 1971 (Sandilands, 1980:60).
11
This has been a chronic problem with low-income public housing even in high-income countries
such as Britain and the United States. In Britain, mainly under Prime Minister Margaret Thatcher in
the 1980s, council house tenants were given the opportunity to buy their homes. Maintenance costs
were thus removed from the public purse. 8
family formations in the rapidly growing cities. On the one hand the supply of savings for
housing finance needed a big boost. On the other hand demand for mortgage finance had to
match a greatly increased potential supply. As the problem was stated by two of the
architects of the new institutional arrangements put in place in 1972, Lauchlin Currie and
Luis Eduardo Rosas (1986):
As rates of inflation rose it became less and less possible, without a heavy subsidy, to
provide an adequate volume of long-term loans at low interest rates. As there were
many competing demands for public expenditure, public low-cost housing was starved
and the economy as a whole was deprived of the large but latent outlet for savings and
the impulse for the generation of activity and employment that building in general could
provide.
To convert housing into a leading sector that could propel the entire economy on to a higher,
sustainable growth path required simultaneous action on both the supply and demand side
of its financing. On the supply side, mortgage institutions needed to give savers safety,
liquidity and a reasonable real return that was attractive relative to the alternatives. On the
demand side, the institutions needed to attract many more borrowers by offering mortgages
on much more affordable terms than before. The conundrum was resolved by focusing on
the problem of inflation.
The ideal would be to eliminate inflation. Then, savers could be offered an attractive interest
rate of, say, 4% in both real and nominal terms, with borrowers then charged, say, 6%. At
this rate of interest, many more borrowers could afford a house compared to a situation in
which inflation is 15% and the nominal interest rate 21% (to give a real rate of 6%), or even if
nominal rate were raised only to 15% with a real rate of zero. For at 21% the interest
payments required in the first year of a mortgage (along with amortizations) would more than
triple; and even at 15% they would more than double. Admittedly borrowers’ real outstanding
debt would decline more rapidly in the inflationary situation, but the initial real cash flow or
“front-end loading” of repayments would be unsupportable for many potential borrowers
whose real incomes would otherwise be adequate.
Unfortunately, Colombia has a poor record on inflation, though not as bad as in other Latin
American countries such as Brazil, Peru, Argentina and Chile where the value of money has
occasionally been destroyed by hyper-inflation of 10,000% a year or more.
12
Much of the
blame lies with lax monetary control. The problem is not merely that inflation has been high
on average, but that high inflation is almost always coupled with greater volatility. This plays
additional havoc with the schedule of real repayments for borrowers if lenders insist on
variable rate mortgages. And if mortgage rates are fixed, long-term lenders have difficulty
attracting funds in the face of the more flexible rates in short-term markets.
3:1 Colombia’s index-linked mortgage finance system
Faced with this difficulty, two other Latin American countries, Brazil and Chile, had earlier
introduced index-linking (or “monetary correction”) schemes, to protect their housing finance
from the distorting effects of inflation. In 1972 Colombia went further than Brazil or Chile, to
make housing (and related infrastructural investment) a motor of accelerated development
while simultaneously protecting the real competitiveness of exports as the second motor (via
regular index-linked adjustments of the exchange rate). What is interesting from the point of
view of lessons for China is that in both Colombia and (as we shall see) Singapore, these
12
The consequential rise in nominal interest rates together with severely negative real rates causes
both supply and demand for long-term finance to collapse. The construction sector bears the brunt of
this – for example in Chile in the early 1970s (Sandilands 1980). The housing market is not immune to
this problem even in developed countries in times of moderate inflation: for example, in the case of
the US Savings and Loan (S&L) debacle of the 1980s. 9
two motors of growth have been viewed as inherently complementary. For without adequate
housing the poorer will be the supply and the health of workers and the higher the cost of
employing them in the export sector. Likewise, housing supply and demand are both greater
with a more dynamic traded goods sector that boosts national income. The important thing is
the greatly increased mobilisation of currently underutilised labour and capital via the hitherto
inadequately explored potential for increasing the real demand for both housing and exports
without recourse to an excessive use of the central bank’s printing press.
3.1.2 The distinction between real and monetary demand and its significance for
housing finance
The distinction between real and monetary sources of demand is sufficiently important for an
understanding of the potential for a non-inflationary expansion of the supply of housing as to
merit a digression at this point. What counts is not demand that is merely backed by an
increase in the money supply. Rather, the key is an increase in real market demand backed
by real incomes and the associated increase in real supply of goods and services in
exchange – together constituting the real market size which was Adam Smith’s key to the
wealth of nations. Smith’s insights were extended by Allyn Young (1928) in his seminal
presidential address to the British Association on “increasing returns and economic
progress,” and by Lauchlin Currie (a Harvard student of Young’s) in a later series of papers
that linked Young’s insights on the possibilities of self-sustaining rather than self-exhausting
growth with his equally deep understanding of monetary theory and policy (Currie, 1974,
1997; and Currie in Sandilands, ed., 2004).
Unfortunately there is much confusion, even among economists, on the monetary causes of
inflation, partly because of semantics when terms such as “bank deposits”, “money” and
“credit” are not carefully defined or differentiated. The money supply comprises notes and
coin in circulation plus bank deposits. However, there are two distinct types of deposit: (i)
highly active balances in demand or current accounts held by businesses and individuals to
bridge the relatively short intervals between current incomes and outgoings, and on which
interest is not paid or is paid at a much lower rate than on the second type of deposit,
namely (ii) the much less active saving or time deposits whose motivation arises from a
different source – the desire for interest income.
Since demand deposits are almost universally considered as a component of money or the
means of payment, it is perhaps understandable why many people include all bank deposits
(including saving deposits) in the definition of money, and even go on to include saving
deposits in non-bank financial institutions such as savings and loan associations and other
mortgage “banks”. However, consider the following scenarios in two consecutive years:
(i) A 10% increase in the actual circulating means of payment – in the sense of notes
and coin plus current accounts (the conventional M1 definition) – used in settlement
of debts between buyers and sellers, when real GDP is growing at, say, 7%. Then
inflation would run at around 3% if demand for money as a percentage of GDP (the
inverse of the velocity of circulation) is fairly constant. At the same time, assume a
similar 10% increase in time or saving deposits in banks and non-bank financial
intermediaries. The inclusion of saving deposits would give rise to a broad “M2” (if
only bank deposits are included) or a broader “M3” definition of money (if deposits in
non-bank financial intermediaries are also included) and this would then grow at the
same 10% rate as the M1 sub-set.
(ii) A 10% increase in M1 but an increase in the growth of time deposits from 10% to
20% so that in this year M2 and M3 now grow at, say, 15%.
If we focus on M1 there should be little or no difference in the rate of inflation (for a given
growth of GDP and velocity of circulation) in the two years. But if we focus instead on the
15% growth of M2 or M3 we may be led to the anomalous and absurd conclusion that a 10
growth in savings is inflationary.
13
For example, how could we then explain, without careful
examination of the composition of the deposits, the very different impact on price inflation if,
in year 2, the growth of M2 were still 15% but were now the weighted average of 20%
increase in M1 (brought about by an increase in central bank reserves) while the growth of
saving deposits is unchanged at 10%?
I have belaboured this point because when in Colombia the new index-linked housing
finance system, launched in 1972, led to a substantial increase in saving deposits, its critics
– who included bankers exposed to new competition and central bankers anxious to shift the
blame for their own failure to control the money supply14
– pointed to the accelerated
increase in overall deposits (“M3”) and blamed the new system – and the increase in
building that it financed – for the increase in inflation.
15
Actually, the initial natural lag
between a very big increase in deposits in the new system and their disbursement meant
that a portion of the deposits were temporarily congealed in a special account in the central
bank. This helped restrain the excessive expansion of the money supply over this period.
When investment is financed by savings out of real incomes backed by savers’ real
production, those resources are available to satisfy borrowers’ instead of savers’ demands.
For example, the resources can now be used in exchange for housing (via mortgage finance
intermediated by the relevant institutions) instead of consumer goods. There is a change in
the composition but not the total of real expenditures. When increased savings are used to
finance new housing, they buy the wages and related consumption of construction workers
as well as the local building materials. The latter have relatively low import content, thus
generating high domestic multiplier effects. Building also employs many unskilled workers
(who nevertheless can quickly learn skills on the job) who were previously underemployed.
The net effect is a potentially self-sustaining and non-inflationary boost to the real rate of
economic growth, along the lines described above.
All depends, however, on real effective demand for housing matching the potential growth of
real saving instead of being reliant on inflationary finance. In fact we have explained how
inflationary finance (that is, unsupported by real savings) repressed demand for long-term
mortgages because of the distortions caused by high and variable interest rates in the face
of inflation. The purpose of the 1972 reforms was to resolve this problem.
3.1.3 The main features of Colombia’s index-linked housing finance system
The key institutional reform was to create a number of competitive, private Savings and
Housing Corporations (Corporaciones de Ahorro y Vivienda – CAVs) whose assets
13
Note too that if we focus on the growth of “credit” (or the total of loans) in the economy we would
have to include loans based on all deposits in the system. But this aggregate can and does differ
widely from the growth of the means of payment (M1), depending on the rate of saving. Thus at one
moment a given growth of credit can be inflationary and at another time deflationary.
14
The central bank was not given independence until 1991. It was often dominated by spending
ministers who treated it as a development bank (using inflationary finance) as well as a central bank
whose main objective should be monetary control. These conflicting objectives can help explain why
Leonardo Villar (2009) finds that monetary policy in Colombia has often failed to be contra-cyclical.
15
See, for example, Miguel Urrutia and Olga Marcel Namen (2011). Urrutia had been on the board of
the central bank in 1972-74 and headed the National Planning Department (DNP) with a change of
government that was unenthusiastic about housing as a leading sector. The modifications he and
subsequent planners made are described with little complaint in Urrutia and Namen’s 2011 paper
(though they did acknowledge the political influence of middle-class borrowers) despite their
debilitating effects. In similar vein, a later DNP head, Juan Carlos Echeverry (2002) is representative
of those who regard housing as relatively “unproductive” without seeing the irony in their view of
investment as “productive” when, for example, it is for bricks, cement, glass and high-speed lifts that
would not be produced in absence of final demand. I attach my review of this book as an Appendix. 11
(exclusively construction and mortgage loans) and liabilities (liquid saving deposits) were
authorised to be expressed in an inflation-adjusted unit known as “units of constant
purchasing power” (Spanish acronym: UPAC), with daily adjustments to the index in line with
a moving average of the previous year’s consumer price index.
This had two consequences. Firstly, by offering savers for the first time a guaranteed real
return (initially set at around 5% net) the CAVs experienced a dramatic inflow of funds that
greatly increased the overall saving rate (and was not simply a diversion of funds from other
institutions) and the proportion going to finance new building and their final owners.
Secondly, on the demand side, even though the initial real rate of interest was set at around
8%, this was only about half of what borrowers had to pay in interest in the early years of a
traditional mortgage contract at that time. Furthermore, the attractive rates made it less risky
for the CAVs to extend the typical mortgage period to 15 years. So demand initially ran well
ahead of the increased savings.
16
This explains why initially demand had to be restrained by
a high (but still affordable) real interest rate and relatively high down payments, and why the
mortgage term was not extended further to 20-30 years as is common in other countries.
The success of the system in boosting housing finance and the rate of house-building to well
over 20% a year, confirmed the hypothesis that savers and borrowers would respond
positively to incentives, and that private builders would likewise respond positively to
increased profits from increased real demand for the finished product. Urban unemployment
fell, and GDP growth doubled to 7%.
This was partly due to the contemporaneous increase in non-traditional exports which also
responded positively to the incentive of a new “crawling-peg” exchange rate adjusted
regularly in line with inflation to avoid overvaluation. The record also showed, however, that
there was no necessary conflict between growth of urban house-building and growth of
urban industry. Indeed the record suggested that they are complementary so long as the two
sectors are assured of an increase in real demand (foreign demand in the case of exports) to
which increased supply can respond without the fear of bankruptcy from a fall in price faster
than the fall in costs that can usually accompany an increase in physical scale.
Unfortunately, again as mentioned above, the system’s dynamism was diluted over time by a
variety of modifications (called “reforms”) that can be explained in two main ways. First there
were the politically inspired complaints by the vested interests of borrowers – usually the
relatively rich – who suffered from money illusion when their outstanding mortgage debts
were adjusted in line with inflation. This was normally at a rate below the rise in nominal
incomes that accompanied the rise in real per capita GDP, but they complained that they
were being exploited by savers, many of whom were small savers much poorer than the
borrowers. Secondly there were those who blamed the system for being inflationary when in
fact the opposite was the case. The system adjusted only to past, recorded rates of inflation,
not to future expectations.
These changes consequently took two main forms. First the index base for adjusting the
system’s assets and liabilities was changed so it progressively fell below the actual rate of
inflation, thus harming savers and benefiting borrowers. Secondly, the CAVs’ savings
16
As mentioned above, disbursements followed savings with a lag. Meanwhile the excess was held
on deposit in the central bank, so the increased saving had a restraining effect on the unfortunate
contemporary inflationary expansion of money (that Urrutia and Nemen [2011] do not mention). A vital
lesson for China is that investment should be financed only by real saving, except to the extent that
the money supply can be expanded in a non-inflationary way if it is in line with growth of real GDP and
the related growth of real demand for money to hold for imminent current transaction purposes. It is
thus possible both to boost real spending while pursuing a policy of severe monetary restraint. 12
deposits were treated as inflationary money balances and subjected to increased reserve
requirements by the central bank. The result was a decline in saving.
A further modification was that the private CAVs were increasingly subject to regulations on
proportion of assets that had to be for smaller mortgages on lower-cost housing for poorer,
higher-risk borrowers. Just as the 1977 US Community Reinvestment Act obliged banks to
increase the share of loans going to sub-prime properties – thereby increasing their risks
and contributing to the global crisis of 2008-12 – so similar regulations on the CAVs reduced
their profitability and dynamism and even bankrupted some of them in the aftermath of a
major recession in 1998-2002. The remaining CAVs were then also allowed to diversify their
loan portfolio away from mortgages. They could also offer fixed-rate mortgages on non-
indexed terms, and as inflation fell consistently to single-digit figures for the first time in
decades most new loans were on these terms. However, the banks insisted that for low-
income borrowers their smaller mortgages would continue to be on the index-linked basis
which still involved lower monthly instalments at early stages of the loans.
3.2 A continuing and growing housing deficit. So, for whom should housing be built?
The result of this mixed history since the early 1970s is that although the absolute number of
good-quality middle-class housing (mainly in multifamily medium- and high-rise apartments
on high-price land in favourable city locations) has greatly increased, and has transformed
the appearance of Colombia’s cities, this type of housing has not kept pace with the even
more rapid growth of new family units in the burgeoning cities. The result is growth in the
absolute number and percentage of families living in inadequate, overcrowded
accommodation on poorly serviced, low-priced land – or on higher-priced land but even
more densely packed in, as the only way to keep down the per capita rents.
17
In face of this growing deficit, the government has been pressured into an increased reliance
on direct approaches to the housing crisis. That is to say, on direct subsidies on the rents
and mortgages of poor families, largely administered by the state-run Instituto de Credito
Territorial (ICT), whose name was changed in 1991 to the Instituto Nacional de Vivienda de
Interes Social y Reforma Urbana, INURBE when it was given wider responsibilities to
“adjudicate subsidies to low-income families “for the improvement, construction or
acquisition of vivienda de interes social [social housing]”.
Subsidies of course place a strain on an already strained national budget, and put pressure
on the central bank to monetise fiscal deficits. In this way demand for a large part of the
housing stock is not expanded via real effective demand from savings but rather through an
inflationary monetary demand. By helping to stoke inflation, this contributes to the
discrimination against long-term mortgage finance on affordable terms (absent a
determinedly protected index-linked system), as explained above. Paradoxically, if subsidies
do increase inflation, it is moot whether they would actually yield a net increase in the
housing stock. If, on the other hand, subsidies are paid for out of general taxation (rather
than from land value revenues), this potentially involves disincentives to work and enterprise
and/or the displacement of other public expenditures that may adversely affect economic
growth, hence lower real demand for improved housing for all.
17
According to the Population Reference Bureau (2010), the urban population in Colombia had risen
to 75% by 2005 and the percentage living in slums was estimated at 18% (more than six million
people). Most of the 25% living in rural areas (11 million people) also lived in poorly serviced sub-
standard housing. For China the figures were 195 million people living in urban slums (38% of the
urban population which in turn was 45% of the total population). As in Colombia, a large percentage
of the rural poor in China have very sub-standard housing. 13
After allowing for the effect of subsidies on rents and mortgages or on disposable incomes,
the direct approach to the reduction of the housing deficit asks how housing can be matched
to what people can afford, and building accordingly. But, almost by definition, most people in
developing countries, including Colombia (average PPP income $8,500 per head in 2008)
and China ($6,020), have incomes that are far too low to cover the building costs of anything
except extremely basic shelter. The proportion of the population living on less than US$2 a
day ($730 a year, or perhaps double that in PPP terms) was recently estimated at 26% for
Colombia and 36% for China (Population Reference Bureau 2010). The annual PPP family
income of these groups would be perhaps $6,000 for a 4-person family in Colombia and
$4,500 for a 3-person family in China. If we assume that 33% is the maximum proportion of
family income that can reasonably be devoted to house rents or debt servicing, this gives for
these groups an annual budget of a mere $2,000 for Colombia and $1,500 for China. If
these figures are capitalised at 5%, this would allow an expenditure of $40,000 and $30,000
for new housing units for these 4 or 3 person families in Colombia and China respectively.
With such low budgets available to so many low-income families, massive subsidies would
be required to deal with the housing deficit if it were tackled by building anything other than
new, poorly located slums for these people. In reality, in Colombia as in all developing
countries there is a constraint on the size of the national budget that can be allocated for
housing subsidies. As with education and health, there is clearly a case for subsidising the
housing needs of the very poor. However, unlike in the case of education and health
services that must be produced directly and continuously to serve the need, with housing
there is a potentially more effective indirect way to meet the needs of the poor. That is
because housing, if well built, is a very durable good.
Consider an analogy with the automobile industry. Manufacturers do not build cars for the
poor; they build them for people who can afford the high costs of production. These new cars
are added to the total stock of new and second-hand cars, and purchasers sell their old cars
on the second-hand market to less wealthy individuals (who in turn sell their old cars) at
lower prices depending on age and condition. In this way it is possible to buy perfectly
serviceable cars that are, say, 10 years old, for as little as $2,000 (or even less) toward the
bottom of this filtering-down (or upward escalation) process. At these prices even some very
poor people manage to afford quite good cars. However, it would be impossible to build a
new car for anything like $2,000.
Likewise, in the housing market there is a huge stock of housing of different ages and
conditions. One difference is that conventionally built housing is even more durable than
cars and, with regular maintenance, can last many decades. The greater the effort put into
ensuring that quality homes are built for those who can afford them (and into maximising the
number able to afford them with appropriate mortgage repayment schedules), the more
dynamic and non-inflationary can be the annual rate of growth of the construction sector.
There would then be less need for subsidies; or a given budget for subsidies would be able
to stretch further. Clearly, subsidy programmes (such as the ICT programmes in Colombia)
cannot on their own be expected to be nearly as dynamic as a programme that includes
middle-income housing that is self-financing because fully geared to effective demand.
The other key point is that if newly built houses are then also of a quality that is at least as
good as the average of the existing stock, and at least match the number of new family units
being formed at the same time, then there would be an upgrading of the quality of the stock
and/or rooms available for the poorest families, and their price would be lowered.
Furthermore, this price would be below the cost of building equivalent new accommodation
and space for those at the bottom of the ladder. These would include families recently
arrived from rural areas. They in turn would have released their rural homes (or space in
those homes) for those remaining behind. The escalation or filtration process can have deep
benign ramifications. 14
However, this result depends crucially on the volume of new good quality building matching
population growth in its most economic locations, and unfortunately this has not been
happening. For example, between 1996 and 2008 the average annual number of new
houses financed by the formal financial system was only 42,453, whereas the average
growth of the urban population over this period exceed 500,000 person, or around 100,000
households if the average household had five persons.
18
3.3. Urbanisation and land values
This brings us to consider further not only the efficiency gains from more rapid urbanisation,
but also to the question of a more equitable distribution of those gains. On the side of
efficiency, we have already stressed the importance of greater mobility of labour and capital.
So long as disguisedly unemployed rural labour can find more productive work in the cities,
the country’s GDP is increased and resources more fully employed.
If the growth rate of the formal construction sector can be boosted by more fully exploiting
the potential to convert latent demand for middle-income housing into dynamic effective
demand this would create many more formal sector jobs with higher incomes than migrants
can obtain on what might be called a do-it-yourself, Robinson Crusoe-style of unlicensed,
informal construction work. Higher wages in the formal sector could then finance better
housing than that constructed in barrios de invasion. It would also increase reciprocal
demand for other goods and services (including, but not only, the industrial and service
sector inputs that are purchased by and then assembled by the construction sector itself),
and be a potentially powerful motor of overall growth. The increased scale of the market
would generally enhance gains from specialisation and urban agglomeration economies. All
of this also has a symbiotic relationship with the growth of exports up a dynamic “ladder of
comparative advantage” as resource endowments and relative factor prices alter.
So from the point of view of increased employment and higher wages, urbanisation improves
the distribution of incomes because it helps to make labour relatively more scarce. In
Colombia where rural-urban migration that absorbs the pool of surplus rural labour has
already proceeded much further than in China, upward pressure on wages has been
stronger than in China, but this may now be changing as China (with its relatively slower
overall population growth) powers ahead economically at a faster pace than in Colombia. At
the same time, this very process of demographic shift to the cities comes up against a fixed
supply of land. The pressure of demand for land from increased population, incomes and
economic activity, all of which requires space in which to live and do business, inexorably
drives up the price.
Normally, a rise in price above the cost of production elicits an increased supply that tends to
reverse the price and bring it back in line with costs. The land market, however, is different.
First, the overall supply cannot be increased when demand increases. The supply can only
be transferred from one person or one use to another. Second, land qua land does not have
any resource (labour and capital) costs of production. It is “the free gift of Nature” (the
classical economist David Ricardo’s famous phrase). Thus, when demand increases its
(resource) cost of production does not increase. However, it does have opportunity costs, so
that one use of the same piece of land displaces other uses, and this is one reason why land
commands a price. The other reason is that some plots are more desirable than others
because of intrinsic differences in fertility (important in the case of agricultural land or mines)
or surrounding amenities (infrastructure of all kinds). This influences the strength of demand
18
Sources: Departamento Nacional de Estadistica (DANE), for data on new houses financed:
(http://www.dane.gov.co/daneweb_V09/index.php?option=com_content&view=article&id=87&Itemid=57);
and Population Reference Bureau (2010) for data on urban population growth. 15
for one site relative to another. Then, for the buyer or renter of land the price is a private
cost. For society the price is only a transfer payment. There is no social cost.
Society, however, rather than the individual owner of a particular site, has built and paid for
the infrastructure that makes the land of a city valuable (over and above its natural features
and location relative to climate, natural waterways etc.). The individual owner pays for the
buildings and improvements on the particular site, but these have only a trivial effect by
themselves on the value of the land on which they sit.
This brief excursion into classical land economics is intended merely to highlight one of the
most powerful forces acting against what would otherwise tend to be the equalising forces of
competition and mobility – to the extent that these are allowed free play – during the process
of development. Whereas competition and mobility promote an equalisation of wages and
returns on saving and investment in man-made capital (buildings and equipment), no such
forces operate in the market for land.
19
Land is both fixed and immobile. Competition for its
services does not drive its price down to its (zero) direct and indirect labour costs of
production. Pure scarcity relative to demand determines its price.
The socially efficient use of land requires that it goes to those who can obtain the highest
social value from the labour and capital applied to it. Price then performs a socially useful
rationing function. But insofar as the overall value of land is a reflection of Nature’s bounty or
the collectively provided civic amenities and infrastructure and purchasing power of the
community, its value is a community value. As such there is a powerful ethical as well as
economic case for ensuring that increases in land values consequent on the growth of
national economic activity and population growth be captured by the community as state
revenue instead of relying so much on the taxation of work and productive enterprise.
Nowhere is this more germane than in rapidly urbanising developing countries such as
Colombia and China. It is also highly germane to Singapore (reviewed below) where, even
though it has been defined as an almost completely urbanised country for the best part of a
century, rapid economic growth continues to put strong upward pressure on land values.
In Colombia “valorization taxes” have been imposed, on a piecemeal basis, to capture the
increase in value of lands surrounding major public works such as road widening schemes
where clear localised enhancements could be identified and where political obstacles could
be overcome. But even in the case of the recent major investment in “TransMilenio” – a rapid
bus transit system for Bogota that greatly reduced travel times along its routes – it has been
hard to disentangle the increases in land values associated with this one project from the
multiple contemporaneous influences of new and old economic activity (Rodriguez and
Mojica, 2008). Thus most of the windfall land value increments have remained in private
hands, except insofar as the city authorities already owned some of the land needed for
such projects or could acquire it at pre-development values before auctioning it to
prospective private developers after granting planning permission for developments
stimulated by the new publicly provided infrastructure. (This was common practice in
Singapore using powers under its Land Acquisition Acts; see below).
19
However, as Mason Gaffney (2012) notes (following John Stuart Mill), the rising relative price of land
induces land-saving technical progress that replaces land with capital (for example, in high-rise buildings, or in
fertilizers that boost farmland yields). Gaffney sees this as a natural curb on rising land values, but while these
developments can reduce urban sprawl and land-extensive cultivation, they also drive up the price of urban and
rural land where intensification is occurring, even as it lowers the relative price of peripheral land. (Also see
Gaffney’s [2009] explanation of how taxes on wages and enterprise depress land rents and how therefore a
switch from taxing production to collecting rents would yield as much or more as existing fiscal revenues,
together with a more free and dynamic economy.) 16
It is well-known that this process of granting official permissions to convert land from one use
to new, much higher economic uses that were previously prohibited or uneconomic is open
to much abuse. In Colombia, as in China today, permission to convert agricultural land to
urban uses can release huge repressed land value and yield windfall gains to owners if left
untaxed. Stories abound of corruption induced by these windfalls. In China, peasants’
smallholdings can be subject to compulsory purchase orders at their assessed value as
agricultural land. This prevents the owners from obtaining windfall gains. A case can be
made for this if the windfalls finance projects that benefit the wider community more equally.
However, not if local officials then sell the land to private developers at a price below its
much higher value in urban use, after pocketing bribes. Henderson (2009: 22-23) suggests a
more transparent auctioning of leases would reduce corruption. However, these dramatic
examples of more-or-less overnight windfall gains should not hide the widespread secular
rise in the value of land throughout existing urban space in the face of ever-increasing
demand. Regular revaluations of properties for local property tax purposes (preferably with a
distinction drawn between value of buildings and value of land, with higher rates imposed on
the latter than the former) would go a long way to redistributing community-created values
from the fortunate few private beneficiaries to the community as a whole.
Rising land values are both the consequence of urban growth (varying with the way in which
infrastructure and amenities are concentrated) and of the way that land is rationed to the
highest bidders. Naturally the rich can afford to live in the most desirable, exclusive
neighbourhoods. The poor are forced to cheaper, remoter and less well-serviced lands, or
live on more expensive land but in densely populated overcrowded conditions. In both cases
the tendency is toward an undesirable degree of social segregation that is another
justification for government intervention in the play of the land market. In Colombia we have
seen that subsidising of the rents and mortgages of the poor has played an important, albeit
subsidiary role in the provision of social housing, enabling the poor to live in better conditions
and richer neighbourhoods than would otherwise have been possible. Ideally, the funds for
this social goal would be derived in much greater degree from the public capture of land
value increments (“participación de plusvalía” is the term for an instrument recently
introduced in Colombia for the capture of up to 50% of land values created by public actions:
see Patricia Acosta, 2008) than from the taxation of productive enterprise.
3.4. Urban design
The typical pattern of urban growth in Colombia has largely driven and been driven by
market-determined land values. City design and its pattern of expansion has also of course
been influenced by the interventions of national and city planners in zoning and deciding on
the location of infrastructure and public housing. The latter projects, however, have tended to
follow and reinforce rather than consciously modify market-driven land values. But some
interesting attempts to buck the market trends have been made. Lauchlin Currie, as adviser
to the Bogota and national planning departments over many years has again played a
prominent role here. In a 1967 study for the then mayor of Bogota, Virgilio Barco (later
President of Colombia), he proposed a multi-centred alternative to further development of a
mono-centred city with dispersion that followed the existing lines to the north and west.
These ideas were incorporated into the 1972 national plan by President Misael Pastrana as
a “cities-within-cities” approach to the future growth of Bogota and other major cities, and, as
mentioned above, Currie’s subsequent book (Currie 1976) was the centrepiece of the 1976
United Nations Habitat conference in Vancouver.
The idea, strongly influenced by Singapore, Paris and Moscow was to try to minimise the
dominance of the private car and to minimise transport requirements by creating within each
metropolitan area, separate distinct centres which would be as self-contained as possible,
with an appropriate mix of residences, workplaces and social amenities serving mainly the
residents of each centre, and making each centre as walkable, hence as densely occupied 17
as possible, yet with ample open spaces partly by minimising road space and the need for
private cars (see Sandilands 1990, chapter 11 for details). The new city of El Salitre near to
but distinct from the traditional centre of Bogota was an example of the partial influence of
these ideas. It was an identifiably distinct centre with a concentration of government offices
and cross-subsidisation of residents that attempted to maximise the number of government
workers who also lived in this new city-within-the city. By locating these new cities within the
wider metropolitan area the benefits of agglomeration economies were not lost. This was in
contrast to policies that sought to establish small new cities beyond the ambit of an existing
metropolis. The idea was to tame the metropolitan design so as to make it more efficient and
more liveable, rather than to curb its growth and sacrifice its natural advantages.
4. Comparisons with Singapore
A partial precedent for the model of metropolitan development that Currie had in mind was to
be found in Singapore where he made two fact-finding visits, in 1973 and 1984,
accompanied by senior Colombian planning officials. During the decade preceding his 1973
visit, Singapore had rehoused one-third of its entire population in a series of large, relatively
compact new towns. By the time of his second visit in 1984 (on a trip that followed an official
visit to Beijing, Hong Kong and the booming new city of Shenzhen), three-quarters of
Singapore’s population had been rehoused and her new towns were increasingly self-
contained cities-within-the city.
Singapore’s new housing developments were almost exclusively high-rise apartment blocks
of ten to twenty stories (with much higher blocks accommodating, amongst others, many of
the rich expatriate community working in the much sought-after downtown financial and
commercial districts where extremely high land values rationally dictated even higher
densities for both residential and commercial activities). An unusually high proportion of
national income was diverted to building, yet the country had one of the fastest growth rates
in the world. The proportional share of building grew progressively over the years – a
reflection of the fact that it was a leading sector with a growth rate significantly higher than
the rest of the economy, thus levering up the overall rate of economic growth, alongside a
vibrant export-oriented manufacturing sector that was its complementary motor of growth.
The coexistence of exceptionally rapid growth of housing and exports belied the fears of
those who thought that only industrial sector investment could be a productive leading sector
while believing that investment in housing would represent an unproductive drag on growth.
Singapore has not had to face the challenge of massive potential migration to its cities
resulting from the displacement of labour by new agricultural technology coupled with low
elasticity of demand, as most developing countries such as Colombia and China have faced.
Singapore has been a city state for most of its modern history. However, as Southeast Asia’s
leading entrepôt city, a very large number of its citizens were until fairly recently engaged in
very low-productivity, low paid, labour-intensive work loading and unloading ships in the
overcrowded harbour district, in slum conditions. By the time of self-governance in 1959 and
independence in 1965, Singapore’s leaders realised that these low-productivity entrepôt
activities (whose workers represented challenges and opportunities similar to those posed
by the mass of disguisedly unemployed rural workers in most underdeveloped countries)
offered only limited scope for a rapid acceleration of economic growth. The present report
focuses on the period of major transformation, 1960-90, when Singapore was facing similar
challenges to those faced by China’s major cities today.
Singapore’s manufacturing programme of the early 1960s demanded that new industrial
areas be developed, with extensive new factory and office building. But it also required new
housing estates and towns in which workers could live in less overcrowded conditions and
closer to the new industries. Efforts were made to reserve 20-25 percent of the land in the
new towns for industrial purposes, consistent with a cities-within-the-city design, but this has 18
not been entirely successful since the mix of jobs and residences has never been adequate
and heavy commuting continues. However, this has diminished over the years, particularly
after the early 1980s when the provision of housing near to the huge Jurong industrial
estate, built on reclaimed swampland, was massively increased and the transportation
system improved. The Economic Development Board (1961) and the Jurong Town Council
(1968) were established to build advance factories on cleared sites (mainly in Jurong but
also in other parts of the island), and to let these on attractive terms to foreign industrialists.
4.1 The Housing Development Board
But the first and most important statutory body to be established, in February 1960, was the
Housing Development Board (HDB), because massive investment in public housing was
considered not a liability but a prerequisite to the overall development programme being
planned. The HDB was given wide-ranging powers to acquire land and create self-contained
new towns to house large numbers in high-rise apartments with related public buildings and
services. Until 1974, when an independent Urban Renewal Authority (URA) was established,
the HDB was also responsible for slum clearance and development of prime sites in the
central business district (CBD).
The HDB replaced the old Singapore Improvement Trust (SIT) that had been administered
by the British Colonial Government between 1927-59. During that period the SIT had built
only 27,000 public housing units and these housed only 9 percent of Singapore’s population.
By contrast, between 1960-90 the HDB used its extensive powers and resources to
complete more than 650,000 units (plus a substantial volume of related facilities), and
rehoused more than 85% of the population (Sandilands 1992).
Before 1960 the rate of new housing units constructed, mainly in the private sector, was
falling far short of new family formations, with the result that the already acute problem of
overcrowding was steadily worsening. The HDB initiated its activities with a five-year crash
programme, 1960-65, when completion of public housing units increased from around 1,000
units a year to an average of more than 10,000 a year (with a further 3,000 a year in the
private sector for most of the period up to 1980, and about 5,000 a year after that). During
1960-65 population pressure was growing by about 50,000 persons a year. This fell to less
than 40,000 in the late 1960s and 1970s, yet the public housing programme accelerated to
over 13,000 units a year between 1965-70; to nearly 23,000 a year between 1971-75; to
27,500 a year between 1976-80; and to more than 40,000 a year between 1981-85 at a time
when annual population growth increased to around 50,000 a year following a reversal of the
earlier policy of population control. After 1985 there was some decline in public housing, with
greater emphasis on renovation and remodelling. At the same time there was a big increase
in the completion of higher-quality and larger private-sector homes, at nearly 5,000 a year.
Thus, during this 30-year period the growth of the overall housing stock far outpaced the
growth of the population, permitting a considerable amelioration of the overcrowding problem
and progressive improvement in the quality of the stock.
The initial emphasis by the HDB was on building small, rather basic apartments in high-rise
blocks for rent to low-income families which, despite the low building costs (arising from
scale economies), involved a heavy subsidy element. Rents were based on the premise that
a family should not have to devote more than 20% of its budget to housing. But rich tenants
paid the same rents as poor. Such a policy was clearly not feasible on a large scale, so the
HDB had to establish strict eligibility rules in allocating apartments for rent and these rules
were progressively tightened over the years (Sandilands 1992; Yuen 2007). Beginning in
1964 after the most acute housing needs had been met, the HDB vigorously encouraged
people to buy rather than rent. Thus, at the same time as the eligibility criteria for rental units
were tightened, the eligibility criteria and terms for the sale of apartments were progressively 19
relaxed. In this way there was always adequate real demand to match the expanding supply,
and fiscal and monetary policy could almost always be kept under tight control.
20
The thinking behind this “Home Ownership for the People” programme was that ownership
gives Singaporean citizens a personal stake in the new nation and fosters a sense of
nationhood. It can also be an effective way to increase the purposeful incentive to save (or,
in Singapore’s case, to increase the willingness to accept a very high rate of compulsory
saving). And, critically, it reduces the amount of state subsidy that extensive application of a
low-rent policy would imply.
4.2 The Role of the Central Provident Fund (CPF) in the Financing of Public Housing
In fact the ownership programme experienced a slow start. This was due (i) to the relatively
small number of units offered for sale in the face of continued heavy demand for rental units;
(ii) to the difficulty people had in raising the 20% down-payment requirement; and most
importantly, (iii) to the relative scarcity of funds available for mortgage loans. A tremendous
boost to the home ownership programme came in 1968 when applicants were permitted to
utilise their accumulated pension fund contributions, under the Central Provident Fund (CPF)
system, for both down-payments and monthly mortgage repayments for their HDB flats.
21
The CPF was started in 1955 with employers and employees each contributing 5% of wages
to the fund (but subject to a ceiling). By 1960 net contributions plus interest amounted to
about 1.4% of GDP, almost all of which was invested in government securities. This aided
the financing of development programmes including those of the HDB. Total expenditures on
new housing in 1960 came to 2.9% of GDP, or 1.9% in value-added terms. (The value-
added contribution of the construction sector as a whole came to around 3.4% of GDP.)
22
Rapid expansion of incomes and employment after 1960 boosted CPF savings but it was not
until CPF rates were raised, in a series of steps beginning in 1968, that the increase in CPF
savings began to outpace the HDB building programme. The rates for employers and
employees were each raised to 8% in 1970. In that year CPF revenues exceeded the total of
all public-sector building projects, including factories, offices, schools, and hospitals as well
as housing. The HDB accounted for 71% of those public investments.
Over the next 14 years CPF contribution rates were increased until they each reached a
remarkable peak of 25% of gross wages (subject to a ceiling).
23
These high CPF rates go a
long way to explaining the remarkable rise in the overall national savings rate that peaked at
45% of GDP in 1990. In view of the country’s usual fiscal conservatism, with the public-
sector budget normally in surplus, CPF revenues explain both Singapore’s relatively very low
taxes and why the country could accumulate one of the largest foreign exchange reserves in
the world, as a proportion of GDP, without inflationary consequences.
20
Despite one of the fastest rates of real economic growth in the world, 1960-2011, Singapore has
also had one of the lowest average inflation rates.
21
The mortgage repayment term was progressively extended from a maximum of 15 years before
1970, to 20 years in 1970 and 25 years in 1986, thus increasing affordability and/or permitting the
purchase of increasingly spacious homes on increasingly scarce, hence high-priced, land.
22
Calculations in this section are mainly based on tables constructed by Sandilands (1992).
23
This means, for example, that if the gross wage (before CPF contributions) is S$1,000 a month, the
employer’s wage bill is S$1,250 and the employee’s CPF account is credited with S$500 in total. This
is 40% of the wage of S$1,250 inclusive of the employer’s contribution.
It is interesting that when in 1985-86 the country faced a recession due to a squeeze on exports
(resulting from a government-mandated high wage policy and a strongly appreciating exchange rate),
the CPF rate was sharply cut from 25% to 15% for employers only. This quickly restored the country’s
competitiveness and growth recovered. 20
Thus the CPF has performed two key roles in Singapore’s massive housing and urban
redevelopment programmes, one direct and the other indirect.
The CPF’s indirect role lies in the forced capture of personal savings which are indirectly
channelled to the HDB, Urban Renewal Authority (URA), Jurong Town Council (JTC) and
other statutory boards through the government’s Development Fund (instead of this coming
out of general taxation). The HDB has usually obtained the lion’s share of these funds,
loaned to it at rates of interest below market rates, though normally at a rate that exceeds
the rate of inflation. Borrowers are charged only a slightly higher rate, and the government
gives the HDB a subsidy to cover any cash deficit on its annual operating budget.
The CPF’s direct role is in permitting its members to use their CPF balances to buy their own
homes. Typically the down payment is made through the use of accumulated balances,
while mortgage debt servicing is through the use of current contributions. Since 1981
members were permitted to use their balances to purchase private as well as HDB
housing.
24
The acquisition of real estate wealth is seen as partly replacing the need for
retirement pensions.
By the late 1960s the share of the public and private construction sector as a whole in GDP
(that is, in value added in the assembly of construction materials) had risen to nearly 10%,
with gross expenditures (inclusive of bought-in inputs) between 15-20% of GDP, rising to 25-
30% in some later years. Gross expenditures on residential construction alone were around
8 or 9% of GDP in the late 1960s and for most of the next 20 years. Thus not only was the
rate of growth of construction above average (with housing a major component) but it
became a significant proportion of GDP, thus fulfilling its role as a significant leading sector
or motor of growth, apparently without in any way preventing the growth of an increasingly
diversified export sector from fulfilling a similar (indeed complementary) role. The Singapore
example indicates a strong symbiotic relationship between the two.
4.3 Housing, land ownership, and the distribution of income
We mentioned above the way in which a rapid growth of population and economic activity in
the cities of developing countries increases the demand for the fixed supply of space,
especially in those locations where socially provided amenities are greatest. This puts strong
upward pressure on the value of land in those areas, permitting great fortunes to be made by
private individuals who have made little or no corresponding contribution to the creation of
wealth. Land values are transfer payments that divert part of the nation’s output from
producers to the owners of land.
Fortunately for its citizens, Singapore inherited large areas of public land from the British in
1959, but the Land Acquisition Act of 1966 and its subsequent amendments gave the
government greatly increased powers of compulsory purchase, with compensation reflecting,
at best, pre-existing use values only. By 1990, due largely to HDB land purchases (as well
as substantial land reclamation by public bodies), the state was owner of around 75% of all
of the land of the island.
The provisions of the Land Acquisition Act greatly reduced the HDB’s financial requirements,
and this is reflected in the relatively low selling prices of HDB flats, thereby spreading the
social land value increases (often referred to as “unearned increments” in the literature) to a
wider public. It should be noted, however, that purchasers do not acquire explicit private
property rights in land, over which the HDB retains ownership and control, and which has
24
Not all of the funds can be used for purchase of housing. Some is retained in “Medisave” and
“Edusave” accounts to pay for a proportion of educational and medical expenses. Withdrawals have
also been permitted for the purchase of specified gilt-edged bonds. 21
been reported to have an average value of around 50% of the market price of housing
(Tyabji and Lin 1989). This is one reason why prices of HDB flats continue to diverge from
those of similar units in the private sector, and the price differential is therefore only partly
due to explicit cash subsidies (which are a very small proportion of total HDB operations).
25
Also, unlike private sector housing, HDB flats are sold only on 99-year leases, not freehold.
However, once families have bought their flats at a subsidised price, they benefit from future
increases in the value of the land on which their flats stand, and they can benefit from these
unearned increments when and if they come to sell in the future, except to the degree that
annual property taxes, with regular updating of values, are imposed on owners.
Of equal importance for the absolute increase in the incomes of the poorest groups in
society – even if not for the relative improvement in their incomes – has been the sustained
dynamism of the economy, due in no small part to the sustained dynamism of the housing
sector. For this has ensured more or less full employment throughout the history of
Singapore’s independence, and the country’s very high (forced and voluntary) private and
business savings have been invested productively in so much physical and human capital
and technology as to ensure a steady rise in real wages even for the relatively unskilled. The
reason why the Gini coefficient of relative inequality has improved less rapidly must,
however, be due in no small measure to the way in which overall economic success is still
largely reflected in the retention of spectacularly high land values in private hands. More
could be done to ensure that even more of these pass to the government (and thence to the
people as a whole in public benefits and/or lower taxes on earned incomes) than they do
already. This would also help to reduce speculative holdings of land that have often been a
significant cause of boom-and-bust land and business cycles.
One final comment regarding the sharing of land values: in an effort to minimise the amount
of land given over to the private car, as well as to reduce the problem of congested roads
that would hamper the efficient conduct of business, the government introduced draconian
restrictions on private car ownership in the early 1990s through the “certificate of entitlement”
(COE) scheme. This involved the auctioning each month of licenses to purchase a new car,
with the number of new licences issued determined by the government’s estimate of the
increase in the carrying capacity of the island’s road network (roughly by 1% a year). This
resulted in very big and growing excess demand for COEs that drove the price of cars to
more than four times the world market price. This was effectively a way to sell or rent
valuable land given over to roads to richer Singaporeans so that they, rather than those
without cars, paid for the roads, but also subsidised the improvement of public transport for
all. (In addition the road space in the central business district was also subject to very high
daily charges.)
4.4 The construction sector and exogenous shocks
Singapore is one of the most open economies in the world and is therefore peculiarly subject
to global economic crises that adversely affect its ability to sell to the world, generating
recession, notably during the 1997 Asian crisis and the current protracted 2008-12 crisis.
Occasionally recessions have had home-grown causes, notably in 1985-86. This resulted
from a combination of (i) artificially high wage increases mandated by the Singapore Wages
Council in their misguided view that this would stimulate a faster rate of labour-saving
productivity increases but which, coupled with an informal exchange rate peg to the US
dollar that was experiencing an unusually strong appreciation at that time, instead led to a
25
In fiscal 1985/86 the HDB introduced a new accounting convention in which the difference between
the selling price of HDB flats and their market cost, inclusive of land, is recorded as a government
subsidy. This made it appear that buyers of HDB flats were getting a subsidy of a third to a half of the
purchase price, though given that ownership and control of land remains with the HDB, this may be
an exaggeration. 22
strongly over-valued real exchange rate that jeopardised Singapore’s export
competitiveness, together with (ii) a speculative frenzy of new office and hotel building led to
over-capacity that could not be sold at prices that covered the private sector’s building and
land costs.
These two causes were inter-related in that the loss of export markets led to an exodus of
the foreign workers and business men and women who had been sustaining the demand for
land and accommodation of all kinds, including hotels, private sector housing, and office
space. The leading sectors of exports and construction in those years both led down instead
of up. Fortunately, the government responded quickly by cutting the artificial (hence
unsupportable) increases in wage costs via a 10 percentage point cut in employers’ CPF
contributions (as mentioned above), and there was a rapid recovery both in exports and in
the related demand for space and building.
In other years of crisis for the trade sector, Singapore has been able to rely on its second
leading sector of construction to compensate.
5. Lessons for China
The above accounts of Colombian and Singaporean experiences offer positive and negative
lessons for China in respect of the role of the construction sector in general and residential
construction in particular in boosting a self-sustainable and more equitable rate of economic
growth. We conclude with a brief list.
1. The experience of Colombia and Singapore both demonstrate the immense importance
of a government-assisted boost to the residential housing sector and related
infrastructure and services in the acceleration of their overall economic development.
This is vital for the geographical and occupational mobility of the work force so that
labour productivity can be maximised and the distribution of income improved.
2. Housing and related infrastructure rely much more heavily than other sectors on long-
term finance out of external funds. They are therefore much more heavily affected by
variations in the rate of interest than is shorter-term finance for working capital (much of
which anyway comes from internal ploughback of business profits and depreciation
accounts) or consumer credit (which generally has a lower social value). This is
because the nominal interest rate tends to vary with inflation, but often not immediately
in tandem with it. This then adversely affects the incentive to save if real rates fall, while
also adversely affecting the demand for long-term finance of the higher “front-end
loading” problem for borrowers. It therefore merits special protection.
3. Discrimination against long-term finance must be offset by measures that both
encourage saving and ensure that housing finance gets a share of total savings that
reflects the relatively high private and social return on private and public housing. Both
the voluntary Colombian CAV and the compulsory Singaporean CPF systems –
especially the latter – have done much to boost the supply of savings channelled to
housing. China has copied the Singaporean CPF system (with its close links with the
Housing Development Board) through its own national Housing Provident Fund (HPF)
system introduced in 1994. However, the coverage is more limited, and monthly
employee and employer HPF contributions much more modest than in Singapore’s CPF
system (see Lan Deng et al, 2011). These could be increased significantly as a non-
inflationary additional source of funds, especially if its administration can be made as
efficient as in Singapore.
4. The more that is done to boost voluntary and forced saving, the less the pressure on
inflationary sources of funds. This message is particularly apt today, at a time when 23
inflation has been accelerating in China (and is being partly suppressed via price
controls rather than a tightening of monetary policy).
5. In view of the external social benefits of a well housed population and the desirability of
a good social mix (reduction in social segregation) there is a case for subsidies, though
budgetary constraints limit the role that subsidies can play in ensuring dynamic growth
as compared to institutional reforms that boost non-inflationary private savings.
6. The provision of housing should be matched to real effective demand. The schedule of
long-term mortgage repayments on good-quality new housing can be made more
affordable if inflation and nominal interest rates are kept low and stable or if index-linked
finance can be introduced to overcome the front-end loading problem, and if mortgage
amortisation terms can be lengthened.
7. China should avoid building new housing directly for the very poor because only by
building extremely basic structures can the costs be covered by rents that the very poor
can afford. Thus these programmes would require prohibitive subsidies and would do
little to improve slum conditions on cheap (because poorly located) land.
8. The indirect approach to the solution to the most pressing housing conditions of the very
poor is likely to be more successful. But this requires that China ensures that the
number of conventionally built apartments with minimum space and quality standards is
at least as great as the number of new family units being formed each year in its cities
where the greatest opportunities for higher productivity employment lie. As Currie
(1976:63) explained the potential benefits of the “escalation” or “filtering-down” process:
Elements in the supply-demand problem become not the proportion of the
population unable to afford new housing, but rather whether there are sufficient funds
and physical resources on the one hand and sufficient demand on the other to result
in a sufficient addition to the stock of housing to match the new households required
each year so that all the new housing may be conventionally built in a modern
construction sector, may be well located in relation to employment, and may be well
serviced. This converts the problem into one of available funds and the terms on
which apartments may be rented, and the growth of incomes and effective demand
for apartments, and the question who receives and who pays the transfer payments
arising from the rise in land values…
…A secret of the success of Singapore of supplying conventionally built
apartments for an increasing proportion of the population at astonishingly low
percentages of their income can be found in large-volume, low cost, high-density
(which spreads land costs), modest initial sized apartments, rapidly rising incomes
and finally some subsidy to the whole operation by central government. This
combination has permitted not only slums to be replaced but also a steady upgrading
in space and facility standards.
9. There should be explicit recognition that investment in housing is at least as productive
as any other investment
26
if the potential real effective demand for it can be actualised
via the mobilisation of real non-inflationary savings channelled toward that effective
demand, supplemented by subsidies to poorer families to the extent affordable out of
central and local municipal budgets.
10. The promotion of the construction sector complements industrialisation. If
industrialisation hits a recession due to a downturn in the global economy or because of
26
Contrary to Echeverry [2002] who is discussed in the appendix; or Gao [2010:5] who notes
approvingly that housing in China has been regarded as a “consumption item”, and contrasts
the “productive sector in state investment” with “dead-end housing expenditure.” 24
pressure to revalue the renmimbi, there are non-inflationary ways in which demand can
be further reoriented toward housing and related infrastructure to compensate. This is
especially feasible for a country like China in possession of huge foreign exchange
reserves.
11. Urban design should foster increased density to economise on increasingly scarce,
hence high-value land. This can be helped by tough measures to restrict the growth of
private car ownership, with revenues raised from the taxation of car ownership used to
subsidise better public transport.
12. Increased urbanisation everywhere goes hand-in-hand with rapidly rising urban land
values. This is one of the main sources of increasingly unequal distribution of incomes
and wealth. The socially created increases in these land values should be returned to
the whole community via its recapture by the government on behalf of the community as
a whole that creates these transfer payments. Governments should avoid the opposite
scenario: capture of government at various levels, central and municipal, by the vested
interests of a new landed aristocracy, including through bribery and corruption, as well
as through the leaders’ poor understanding of classical rent theory. 25
Appendix
My review27
of the following book further explains why housing and urban infrastructure
deserve special protection in a strategy to accelerate development, and why it should not be
regarded as “unproductive” compared to other investments.
Juan Carlos Echeverry (2002), Las Claves del Futuro: Economía y Conflicto en Colombia.
(Keys to the Future: Economics and Conflict in Colombia.) Bogota: Editorial Oveja Negra.
This book represents the thought and experience of a prominent economist who directed
Colombia’s National Planning Department (DNP) during the crisis years of 1998-2002. Upon
demitting office with the change of government in August 2002, Juan Carlos Echeverry
assumed the post of Dean of the Faculty of Economics at the prestigious University of the
Andes where he continues to exercise great influence on students and economic policy-
makers. His book surveys the evolution of the Colombian economy over the past 30 years,
with a focus on the last decade during which time Colombia embarked upon an ambitious
programme of “apertura”, or openness.
Echeverry notes that he was a member of the team that negotiated with the IMF, and was
basically sympathetic to the “Washington consensus” that stressed “fiscal health” (smaller
deficits), privatisation, financial liberalisation, central bank independence, and tax, pension,
and labour-market reforms. He laments that the vicious and very costly “narco-guerrilla war”
that Colombia has suffered during this period has greatly offset the benefits of allegedly
sound economic policy, and has given liberalisation an undeserved bad name.
The appearance of Las Claves del Futuro (The Keys to the Future) in Colombia is very
timely. It almost coincided with the publicity surrounding the celebration of the centenary of
the birth of another economist who achieved great prominence as a teacher and practitioner
in Colombia: Lauchlin Currie (1902-93). After a distinguished career at Harvard, the Federal
Reserve Board, and in the White House as Franklin Roosevelt’s economic adviser from
1939-45, Currie came to Colombia as head of a World Bank mission in 1949 (see
Sandilands, 1990, and Laidler and Sandilands, 2002). He was invited to stay on as an
adviser to successive governments for the next 40 years. Most notably he was the father of
a unique index-linked housing finance system (known in Colombia by its Spanish acronym,
UPAC, for “unit of constant purchasing power”) which he stoutly defended from its birth in
1972 until his own death in December 1993.
Echeverry gives considerable space to an analysis of that innovative but highly controversial
housing finance system. He notes that its main original aim was to make construction a
“leading” sector that could permanently boost the overall economic growth rate (as explained
in detail in Currie, 1974). In 2002 the system lay in ruins, the victim of countless debilitating
modifications. Construction had for the past four years been leading the economy down
instead of up. Many thousands had lost their jobs, both directly and as a result of the
depressed state of the industries that supply the construction sector. Urban unemployment
stood at 18 percent, with the official figure for urban underemployment standing at 33
percent. In other words, only half of the workforce was fully employed (rural
underemployment was worse than the urban rate.)
The two directors of the National Planning Department who implemented Currie’s celebrated
Plan de las Cuatro Estrategias (Plan of the Four Strategies) between 1971 and 1974 were
Roberto Arenas and Luis Eduardo Rosas. At the Currie centenary in October 2002 both
recalled the great impulse that UPAC had given to construction and to overall growth in the
27
Published in Journal of Economic Studies, 30:2 (2003), pp.183-7 26
1970s, together with full employment and improved distribution. Comparing the dynamism of
the 1970s with the stagnation of today, Rosas remarked, in a newspaper tribute to Currie:
“!Como nos hace de falta en estos momentos!” (“How we have need of him now!”)
Echeverry concedes that in the 1970s and 1980s construction played a very positive role. It
created jobs and was a contra-cyclical influence. But he claims that in the 1990s, with
“apertura” (the policy of economic openness) and the great influx of external credits and drug
money, construction became a pro-cyclical speculative activity that was bound to collapse,
as collapse it did in 1998 – though he stressed that the problem was compounded by the
fiscal strain of a worsening civil war. His main complaint is that construction is a sector that
has been privileged and has diverted resources from traded goods. It also relies on unskilled
workers whereas the future depends on skills and an allocation of resources more in tune
with the market forces that Adam Smith emphasised. Let us rely on Smith, says Echeverry,
not on protectionist “models of development” and privileged “leading sectors”.
However, Currie’s ideas were also heavily influenced by Smith. Currie’s mentor at Harvard in
the 1920s was Allyn Young, whose presidential address to the British Association in 1928 on
“Increasing Returns and Economic Progress” (Young, 1928) has inspired modern
development theory, to which Currie has been an interesting contributor (see, for example,
Currie, 1997). Echeverry does not refer to this so-called endogenous growth theory, perhaps
because of his profound scepticism of “models of development”. But Young and Currie were
inspired by the opening chapters of The Wealth of Nations where Smith emphasised that the
key to increased productivity was specialization or the division of labour. And the division of
labour in turn depended upon the size of the market, or upon real demand.
Today demand management is generally associated with Keynesian policies to tackle short-
run business cycles around a secular trend. These cycles are closely associated with
interruptions to the flow of monetary incomes and expenditures (or monetary demand). But
Smith and Young focussed on competition, openness, and the mobility of labour to increase
the underlying trend of real demand and market size (or what Smith also called “the power of
exchanging”), hence specialization, hence productivity.
Young explained that in the modern economy specialization takes the form of new, more
specialized firms and industries that compete against the old. They introduce new forms of
organization and technology, but only as and when it pays to do so. The larger the market
size, the greater the incentive to innovate. Currie extended this idea to show that the existing
growth rate (of the overall market, or GDP) had a tendency to perpetuate itself. But in
Colombia where resources were abundant but grossly malallocated and underutilized, and
where labour mobility was very poor, growth fell far short of potential. This self-perpetuating
(or endogenous) growth rate was a vicious circle that could best be broken by institutional
measures to liberate the great potential supply.
This is where Currie’s vision of construction’s potential role goes rather further than
Echeverry’s. As a “leading sector” it is valuable not so much as a contra-cyclical, stabilizing
force (though it could also serve that purpose). Rather, it could help Colombia (and other
countries) to break free of her historically slow, endogenous growth path. It is a leading
sector because (a) it is an important direct and indirect component of GDP, so its growth has
a significant effect on overall growth; (b) it moves independently of movements in the rest of
the economy, and can be moved exogenously through discretionary policies; (c) it plays a
vital role in promoting labour mobility; and (d) it is a sector with enormous latent demand.
But in the past this latent demand had been severely repressed. Potential homeowners can
usually only buy a home with the help of substantial mortgages. Thus effective demand
required a rapidly expanding flow of credit on convenient terms. This was not available
because chronic inflation discriminated against housing finance. High inflation requires high 27
interest rates to attract savings. But for people borrowing large sums high interest rates
impose a severe cash-flow problem in the short term (the so-called “front-end loading
problem”). This curtails effective demand. By contrast, “constant value” savings and loans
made it both more attractive to save and easier to borrow.
Echeverry is a stout opponent of inflationary finance. But in Colombia chronic inflation has
been a reality for the past 50 years. Echeverry fails to highlight the main distortionary effects
that inflation introduces. In practice some sectors suffer far greater harm than others. The
disadvantaged sectors – mainly construction and exports - are not “privileged” when
measures are introduced (such as UPAC and realistic exchange rate policies) that protect
them from the distortions of inflation. Furthermore, conventional macroeconomic policies
designed to squeeze inflation out of the system (through temporarily higher interest rates on
government bonds) and to reduce the fiscal deficit (much of it due to depressed incomes)
can also damage the housing sector by making it less attractive to place savings there.
Currie understood how to combine deflationary monetary and fiscal policies with policies to
reactivate the real economy by redirecting incomes and expenditures toward leading sectors
that rely not on the printing press but upon genuine savings. The great economist Harry G
Johnson similarly distinguished between “expenditure-reducing” and “expenditure-switching”
policies (Johnson, 1958). A blueprint for such a combination, with detailed quantitative
estimates of the size of the required “compensatory” effect required of the leading sectors,
was drawn up by Currie and Alvaro Montenegro (1984) as advice for the administration of
President Belisario Betancur in the mid-1980s. The advice was not taken and the country
went through a very bad bout of instability and capital flight.
The construction sector has another very important role ignored by Echeverry: it is an
indispensable element in the labour mobility mechanism. This promotes not only faster
growth but also better distribution. In Colombia there is still a great imbalance in the
allocation of labour, notably between low-paying agriculture and high-paying urban activities.
And in cities like Bogota there is urgent need for better balance between where people live
and where they work, and for an improvement in the quality of the housing needed and/or
demanded by all income groups.
Echeverry claims there is a conflict between investment in internationally traded goods and
investment in housing. Yet in countries such as Singapore, noted for spectacular export
growth, investment in housing has also been enormous, and far greater than in Colombia.
Despite re-housing almost the entire population in the last 35 years its construction sector
still booms. There has been no saturation of demand.
As people’s incomes have increased so Singaporeans have demanded better and better
accommodation and related infrastructure. This has been aided by low inflation and large
pension fund contributions that are released for housing finance at low rates of interest (but
positive in real terms). Building is concentrated on well-built conventional high-rise blocks for
the middle classes. As these families move into new homes lower-income families move into
the ones they have vacated. This “filtration” or “escalation” process enables poorer families
to enjoy far better accommodation than the type of subsidised “vivienda de interes social”
(popular housing) that Colombia is desperately trying to provide today out of limited fiscal
resources.
Housing and exports are complements, not substitutes. Both are capable of expanding on
the basis of a stimulus to and redirection of real savings, rather than via inflationary finance
or subsidies. Here are the real “keys to the future”.
By contrast, Echeverry’s overview of the Colombian economy and economic policies during
the last 30 years focuses mainly on the structure and balance of the national budget and the 28
rate of growth of money and credit. This is rather typical too of the focus of the international
lending agencies when drawing up conditions for further foreign loans. There is much of
value in this analysis. However, its key limitation is that it gives too little weight to the
dynamic changes in the composition of real incomes and expenditures over time in
developing countries like Colombia, and of the need to ensure that the country’s abundant
natural and human resources are allocated – and reallocated – accordingly. Herein the
supreme importance of the mobility mechanism, and of the related role of a dynamic and
well-funded construction sector.
References
Acosta, Patricia (2008), Policy Learning: New challenges for smart value capture in
Colombia. MSc dissertation, Massachusetts Institute of Technology, Boston.
Choi, S. (2000), “Agenda for China’s Urbanisation Policy: Economic Mobility and
Integration”, World Bank urbanisation workshop, Beijing, and in Urbanisation of China:
Patterns and Policies, World Bank, Beijing, 2002.
Currie, Lauchlin (1974), “The ‘leading sector’ model of growth in developing countries”,
Journal of Economic Studies, new series 1:1, 1-14.
Currie, Lauchlin (1976) Taming the Megalopolis: A Design for Urban Growth. Oxford:
Pergamon Press.
Currie, Lauchlin (1997), “Implications of an endogenous theory of growth in Allyn Young’s
macroeconomic concept of increasing returns”, History of Political Economy, 29:3, 413-43.
Currie, Lauchlin and Alvaro Montenegro (1984), Crecimiento con Estabilidad: Un Modelo.
Bogota: Fundación Simon Bolivar.
Currie, Lauchlin and Luis Eduardo Rosas (1986), UPAC: A Theory Converted into
Successful Reality. Bogota, Instituto Colombiano de Ahorro y Vivienda.
Echeverry, Juan Carlos, Las Claves del Futuro: Economía y Conflicto en Colombia. Bogota:
Editorial Oveja Negra.
Gaffney, Mason (2009), “The hidden taxable capacity of land: enough and to spare,
International Journal of Social Economics, 36:4, 328-411.
Gaffney Mason (2012), Georgist Journal.
Gao, Lu (2010), Achievements and Challenges: 30 years of housing reforms in the People’s
Republic of China. Manila: Asian Development Bank Working Paper Series No.198.
George, Henry (1897), The Science of Political Economy. New York: Robert Schalkenbach
Foundation, 1953 edition.
Henderson, J. Vernon (2009), Urbanisation in China: Policy issues and options. Mimeo,
Brown University: China Economic Research and Advisory Program, pp.30.
International Labour Office (1970), Toward Full Employment. ILO, Geneva.
29
Johnson, Harry G (1958), “Towards a general theory of the balance of payments,”
International Trade and Economic Growth: Studies in Pure Theory. London: Unwin
University Books, pp.153-68.
Laidler, David and Roger Sandilands (2002), “An early Harvard memorandum on anti-
depression policy: An introduction,” History of Political Economy, 34:3, pp.515-52.
Lan Deng, Qingyun Shen and Lin Wang (2011), “The Emerging Housing and Policy
Framework in China”, Journal of Planning Literature, Sage publishing on-line.
Population Reference Bureau (2010), Washington, DC. www.prb.org.
Rodriguez, Daniel A. and Carlos H. Mojica (2008), “Land Value Impacts of Bus Rapid
Transit: The case of Bogotá’s TransMilenio.” Lincoln Institute of Land Policy, Land Lines
(April).
Sandilands, R. J. (1980), Monetary Correction and Housing Finance in Colombia Brazil and
Chile. Westmead, UK: Gower Publishing Company.
Sandilands, R.J. (1990), The Life and Political Economy of Lauchlin Currie: New Dealer,
Presidential Adviser, and Development Economist. Durham NC: Duke University Press.
Sandilands, R. J. (1992) "Savings, Investment and Housing in Singapore's Growth, 1965-90",
Savings and Development, XVI:2, pp.119-44.
Sandilands, R. J. (ed.) (2004), New Light on Lauchlin Currie’s Monetary Economics in the New
Deal and Beyond. Special issue, Journal of Economic Studies, Vol.31, Nos. 3/4, 170-403.
Sandilands, R.J. (2009), “Solovian and New Growth Theory from the Perspective of Allyn
Young on Macroeconomic Increasing Returns”, History of Political Economy, 41 (annual
supplement), pp.285-303.
Tyabji, Amina and Lin Kuo Chin (1989), “Financing of Public Housing in Singapore”,
Southeast Asian Journal of Social Science, 17:1.
Urrutia, Miguel and Namen, Olga Marcela (2011, Historia del crédito hipotecario en
Colombia, Bogotá: Documentos CEDE, University of the Andes.
Villar, L. (2009), Governance, Transparency and Accountability in Colombian Central Bank
and Financial Regulation. New York: Columbia University meeting on central bank
governance, April 27-28. (http://policydialogue.org/files/events/Villar_governance.pdf)
World Bank (2009), World Development Report 2009: Reshaping Economic Geography.
Washington DC: World Bank.
World Bank (2012), China 2030: Building a Modern, harmonious, and Creative High-Income
Society. Washington DC: World Bank.
Yuen, Belinda (2007), “Squatters No More: Singapore’s Social Housing,” Global Urban
Development, 3:1, pp.1-22.
Young, Allyn A (1928), “Increasing returns and economic progress,” The Economic Journal,
38 (December), pp.527-42.
1 comentario:
Hey, nice site you have here! Keep up the excellent work!
High Rise Apartments for Sale
Publicar un comentario