to ground realities
Reza Ali, the
urban planner, has conducted a path-breaking research and published the
findings of that research in a paper titled “Estimating Urbanization”.
The paper, that was published in the Political Economy pages of The News on
Sunday three weeks back, does something startling in the context of
Pakistan. Of course he and his team waited a few years to see if a new
census was conducted. Thinking that as an improbability, he relied on the
data from 1998 census because all he wanted was to get a sense “of the
magnitude and trends of the urbanisation phenomenon and map these accurately
across the geographical space of Pakistan.”
To begin with, his paper
states that the categorisation of urban and rural is inadequate and
imprecise. Unlike the previous method, in this work he defines both urban
and rural. There seems no reason, he says, to segment the population into
just these two categories; therefore he brings in the concept of an
‘urbanising’ area or areas in transition, which do not fall in either
rural or urban definitions.
According to this new
categorisation, all of Balochistan except Quetta, large parts of Sindh, KPK,
and the Seraiki sub-province in Punjab are rural.
Of course that is not
where the research stops. It goes on to shatter various myths, including the
dominance of both the ‘rural first’ lobby as well as those that sell
‘cities as the engines of growth’. But such an impressive research must
lead to policy implications that should benefit the people of this country
in terms of resource allocation and reducing inequality. In today’s
Special Report, besides asking Reza Ali himself of what he makes out of this
research, we have asked a different set of economists, academics and
planners of what kind of impact is it going to make in the months and years
As for Reza Ali himself,
with the social, economic and political inequalities being at the level that
they are in Pakistan, the question worth asking is not “Can Pakistan
survive?” It could well be “Should Pakistan survive?” This indeed is a
tough question for us as a nation and as individuals.
In early August,
in an article entitled, “The Urban Present”, I discussed the work of
Reza Ali, an urban planner and researcher from Lahore, who wrote a
path-breaking article in 1999, called “How urban is Pakistan?,” in which
he argued that perhaps half of Pakistan was ‘urban’. The short newspaper
article led to much debate and discussion in academic circles and the
impression many of us had of Pakistan being urban then, was backed up by
very thorough research provided by him.
In many ways, it was a
definitive essay which fundamentally changed how the urban/rural divide is
seen by scholars in Pakistan. He has followed up with his previous detailed
research and published a follow-up article in The News on Sunday entitled,
“Estimating Urbanisation”, and makes an even stronger case for ending
the debate, once and for all, that Pakistan is predominantly a rural
His argument is based on
many key points. Firstly, definitions matter. On the one hand, ‘rural’
and ‘urban’ seem clear terms with contrasting images: isolated farms,
tiny hamlets, cultivated fields, villages, versus the thriving city,
skyscrapers and slums. This may have been a simple way of defining
‘urban’ and ‘rural’ some centuries or even decades ago; this
dichotomy is comfortable but imprecise, over-simplified.
It is no longer a single
homogenous activity, but is multi-functional and diverse. The urban/rural
divide appears as a gradient, rather than a dichotomy. There does not appear
to be a natural dividing line or break point between rural and urban areas
any more. Many social, cultural, economic and environmental issues are
inadequately addressed by current approaches separating ‘rural’ and
‘urban’ agendas. Behaviour and conditions change drastically along the
gradient, but there seem no compelling reason to segment into just these two
Reza Ali makes the
argument that there is no reason to restrict analysis to just these two
categories — urban and rural — and one can introduce the concept of an
‘urbanising area’, which is an area that does not meet the criteria of
an urbanised area as we defined, yet, it has both an urban core and an
overall density higher than that for a rural area. Thus, it’s clearly not
rural, but, it has not urbanised yet, hence the term ‘urbanising’.
The population which has
physically not moved to the cities has adopted urbanism as a way of life,
reflected in changing pattern of consumption and use of services. The
cumulative effect of this has been intense urbanisation, city populations
are much higher than what official data is prepared to reflect, and there is
a connectivity and integration of services and manufacturing access across
This redefinition and
remapping of the urban and the rural has made Reza Ali argue that Pakistan
is evolving a system of cities, and developing urban regions — connecting,
linking, integrating trade, services, manufacturing, and the work force —
within city core and suburbs, peri-urban areas, satellites, small towns and
neighbouring villages, causing the co-movement of urbanisation and
informality, a phenomenon seen in many other global cities.
What Reza Ali’s articles
do is that they lay out the fact and nature of urbanisation and of the
urban, they do no traverse into the direction of what this means. That is
left to the reader. Clearly, one must attempt to make something of this
information and interpret it in order to give further value to his work.
There are numerous
outcomes and consequences from his research which are fairly obvious.
Firstly, and perhaps most importantly, there is not just an urgent need for
a census, which is five years over due to be undertaken, but when the new
census takes place, there is a need to critically rethink definitions and
categories of the urban and the rural. As Reza Ali shows, definitions matter
fundamentally to how we map a population, hence, there is a need to be able
to capture the nature of population dispersal spatially in a way that it
would make some real sense, for whatever purpose it is meant to serve.
policies, and plans would need to accommodate the real reality which exists,
rather than some mythical ideas of rural villages and villagers, most of
whom now live in cities or live in areas designated as ‘rural’, but
dependent on non-agricultural incomes.
This would also mean that
electoral boundaries need a sharp remapping as well, as do administrative
units, having implications for, especially, local government. There is no
way any of the provincial governments can go back to Ziaul Haq’s local
bodies system with a dividing line between urban and rural.
What is required is a
continuum in how units are designed, reflecting how the population lives and
the nature of services they require. Ironically, Musharraf’s district
government system is a better reflection of the spatial layout than the
other systems dividing urban and rural, and although it needs to be
considerably reformed, it does reflect reality more than other options being
Having taught Pakistan’s
economics for three decades, the most frustrating set of statements which
students repeat ad nauseam, through no fault of their own since the
government and others only confirm them, are the following: Pakistan is a
rural country, Pakistan is an agricultural country, and that Pakistan is
‘feudal’. Anyone who reads Reza Ali’s two articles would desist from
making such nonsensical statements once and for all.
The writer is a political
increasingly an urban country. According to the latest Economic Survey,
Pakistan is almost 38 per cent urban in 2013, with projections of 50 per
cent by 2030. Reza Ali is challenging this story. Official figures seem to
hide more than explain Pakistan’s true urban picture by dividing the
country into water-tight rural and urban areas.
Ali provides the missing
link by introducing the concept of ‘urbanising’, defined as areas which
clearly are not rural and have an overall higher population density and
proximity to an urban core. In other words, it is an area in transition.
Based on this new concept,
lumping together urbanising and urban areas gives a different picture.
Pakistan was almost 47 per cent urban in 1998 on the basis of his estimates
and, most certainly, much more now. Our own preliminary projection, based on
Reza Ali’s initial estimate and assuming the same growth rate as implied
in official projections, is that the urban population will be around 70 per
cent of the total population by 2030.
Another dimension of his
estimates is that official numbers are overestimating urban population in
Sindh and Balochistan and understating in case of Punjab and Khyber
Pakhtunkwa. It is Reza Ali’s contribution which makes us think that
urbanisation process is of varying degrees and intensity across Pakistan.
Further, it is not appropriate to divide the country in urban and rural
boxes. There are suburban areas which may be considered urban on the basis
of population density and proximity but enjoying fewer urban facilities and
opportunities as in an urban core.
Looking at urbanisation as
a process has implications for the way we look at inequality. There is a
significant and positive relationship between “social goods” and
equality. In this case, equality is not same income but relative access to
social goods, which appears in income, wealth, and status.
A person or society can
benefit at the expense of others and seize the opportunity of growth. It may
emerge from natural and moral differences. Differences in wealth, power,
status or class are moral inequalities. Differences in physical and mental
abilities, age, health, strength and intelligence are all physical
inequalities. Fair access to health and education create a more productive,
just and enabling society.
By way of common
observation that people move to urban residence because they think there is
better access to civic amenities. They also perceive to be treated more
justly and feel more equal in terms of economic, social and political
Urbanisation has been
advocated on the parameters of equality. Here equality means equal worth of
all citizens. An equal society is a just society where no one can influence
resource allocation and the right of equal citizenship. It does not
necessarily require equality in income.
According to John Rawls,
an American Philosopher (1921-2002), inequality creates unequal access to
the political system and to positions of power. In this way, extreme
inequality undermines democracy. Inequalities are socially undesirable
because it causes inequalities of opportunities. In 1955, Simon Kuznet, a
Russian-American economist (1901-1985) focused on the process of
“urbanisation which is followed by industrialisation” as a means to
achieve equality of opportunity. But it entailed relatively higher
productivity as well as inequality due to the social structure.
hypothesis explains the spatial distribution of economic development in
which individual pursuit of self-interest increases income inequalities
which can generate all types of health and social problems because of
rent-seeking, corruption and expropriation.
inequalities go hand in hand. The Gini coefficient, the simplest measure of
inequality, was 0.35 in urban areas and 0.25 in rural areas in 2005-06.
Rural population is, thus, less unequal in Pakistan. But this tells only
half of the story. First, we are discussing inequality only in terms of
consumption expenditure and not considering moral inequality in Pakistan.
underestimating urban population we are understating the extent of
inequality. Thirdly, a more powerful indicator of rural inequality is the
trend of concentration of land ownership. The Gini in this regard, has been
rising, leading to increasing landlessness.
Pakistan is increasingly
an unequal society. Low-ranking in terms of Human Development Index and
failure to achieve MDGs are pointers in this direction. Another way is to
look at the effect of government action in education and health in
urbanising areas. A specific example is net primary enrollment rate and full
immunisation in four provinces of Pakistan.
In the rapidly urbanising
area, which is Punjab and KPK, the demand for full immunisation as a social
outcome is higher as compared to Sindh and Balochistan. It is understandable
for Balochistan but not for Sindh as it has the second largest urban
population. Net primary enrollment follows roughly the same trend.
needs to be looked at from a fresh perspective. It has far-going
implications for what we know about inequality in its multifarious
The writer is Head of
Division, Finance & Business Economics, University of Central Punjab
Business School, Lahore
“Out of the
petty rural-urban squabble,
The News on
Sunday: In your paper, Estimating Urbanisation, you have concluded that the
categorisation of rural and urban is inadequate and imprecise. To begin
with, just give us a sense of the process of arriving at a definition, what
constitutes a better definition, and is one country’s definition good for
Reza Ali: The
categorisation of the rural and the urban is deeply ingrained: be it our
vernacular socio-cultural idiom, our folk cosmology, our literature and
poetry, ‘putli-ka-tamashas’ (puppetry), theatre and cinema, the
‘sheri-babu’ a’la Dilip Kumar or the Sharif-Singh ‘dehati-aurat’,
we have a good, seemingly clear idea of what we imply when we speak of the
‘urban’ or the ‘rural’.
You, of course, are
referring to the discussion in ‘Estimating Urbanisation’ where the
concern is with measurable, verifiable indicators. You see the census
defines ‘urban’, and then proceeds to classify as rural everything other
than what they term urban. In Pakistan, the definition of urban used since
1861 included all municipalities, civil lines, cantonments, and every other
continuous settlement of at least 5,000 persons that the census commissioner
decided “to treat as urban for census purposes”. The 1981 census adopted
an administrative definition treating as urban only the population within
the boundaries of what were towns, municipalities or cantonments. G. M. Arif
has estimated that considering the population of 361 places of over 5,000
population and urban characteristics better that many places
administratively classified as urban, adds 6.5 per cent to the urban
population of the country — this means that almost 40 per cent of
Pakistan’s 1998 population was urban.
So, there are the varying
definitions, and the definition changes, that complicate comparisons over
time; this should not be interpreted to mean that national definitions are
flawed, but that they distort inter-country comparisons. Further that in
many instances, the implied urban/rural dichotomy is inadequate to reflect
the degree of agglomeration.
TNS: Tell us about your
methodology and the data. If you have relied on the 1998 census, how much of
what you are saying now is relevant and useful because this data may have
RA: We began working on
our Urbanisation Research Programme when the results of the 1981 census were
published and our early work was based on the 1981 data. Once the
preliminary results of the 1998 census were released, we began a detailed
analysis of the urban census results over the entire century, 1901 to 1998.
In the course of this work, we produced a compilation of 100 year census
data; published an urban places map and data-sheet based on the 1998 census;
carried out field research — of course, given the fact that as our entire
work is self-financed, we have resource constraints; conducted a study of
the effects of the migration at Independence on our settlement pattern; and
in 2011, we presented our work on rurality, ‘Pakistan the Rural’.
Earlier this summer, we made a presentation ‘Pakistan — an emerging
You ask about our use of
the 1998 census data, which is the latest data available, and how relevant
it would be today — our purpose is to provide an understanding of the
nature of the urbanisation process, using the latest numbers as an
indication of the magnitude and trends of the urbanisation phenomenon and
map these accurately across the geographical space of Pakistan. This
provides a good basis for future analysis and change, once, and if, a new
census is taken.
TNS: You have brought in
an additional category of urbanising areas. Please explain what is that?
Isn’t there a need to redefine the rural instead of calling it urbanising?
RA: The census defines
only the urban, classifying all the rest as rural. In our work, we define
both, urban and rural, on the basis of measurable indicators of population
density, urban core and proximity to city, rather than administrative
definitions. The approach was proposed by Uchida and Nelson, based on
earlier work by Chomitz, Buys and Thomas, and used in the World Bank’s
World Development Report 2009. This allows us to adequately capture human
settlement concentration. We have adopted significantly higher thresholds
for the urban than those used in the earlier work; thus we classify all of
Balochistan except Quetta, large parts of Sindh, Pakthunkwa, and the Seraiki
sub-province in Punjab as rural.
The framework leaves a gap
between what we are considering rural areas and urban areas. As we show, the
categorisation of rural and urban appears inadequate and imprecise and there
does not appear to be a natural dividing line or break point between the
two: the urban/rural divide appears as a gradient, rather than a dichotomy
— life changes in a variety of dimensions along the urban-rural route:
from fields and intensive cultivation, villages and small market towns, to
larger towns, small cities and the cosmopolitan city and is not a single
homogenous activity — it is multi-functional and diverse. There seems no
compelling reason to segment them into just these two categories.
We introduce the concept
of an ‘urbanising’ area for areas which clearly are not rural since they
have both an urban core and an overall density higher than the criteria we
are using to classify the rural but do not achieve our criteria for urban
areas — they could be considered in transition.
TNS: How does this new
categorization help in matter of policy; you have mentioned certain areas
line poverty, politics, gender, development, empowerment, governance,
RA: I am merely a
researcher and must not venture into the realm of policy. But obviously it
does have implications, some rather of a fundamental nature.
On the one hand there is
the ‘rural first’ lobby where zamindars lobby for subsidies on
agricultural inputs — water, tubewells, energy, fertilizers, and more
justifiably, investment in rural areas, which is fine as people live in
these areas and are entitled to services, facilities and the benefits of
development; this lobby resonated well with the bureaucracy and the
traditional political masters.
On the other hand, there
is the power of new money, and its hand-maiden, the middle class and the
chattering crowd, concentrated in large cities. They provide the steam for
the engines of growth that cities are supposed to be — so, we are told
cities is where the investment must go, policy must focus on — the
pendulum is now to swing in the other direction, everything must happen in
the cities. Once again, realistically speaking, not only do people of the
cities deserve services, but also that cities are the easiest to provide
with infrastructure and services, there are large economies of scale. The
provision of municipal services, transport, electricity and communication
are the easiest in cities and I dare say, if you can’t do it in cities,
then you really don’t have it in you to do it anywhere else either. Every
rupee spent in the city benefits much more people than it would in rural
We cannot take the need
for rural development to the level of a religion in itself, nor preach the
mythology of cities as engines of growth to the exclusion of everything
Cropping and livestock
directly account for a little less than a quarter of our GDP, directly
employ 40 per cent of the work force in addition to those in the supply
chain, transport, etc.; agricultural exports constitute more than 11 per
cent of the total and downstream industries, textiles add another 40 per
cent. Yes, agriculture’s share is down from almost 46 per cent in 1960,
but really it’s not just the large cities that sustain the economy of
Pakistan. Actually, the rural areas do provide the resilience to our economy
and our ability to bounce back after every natural and man-made disaster
that we have faced can largely be attributed to it.
TNS: So is that the
problem with the thesis “cities are the engines of growth”?
RA: Personally I feel
it’s looking at things in black and white that’s the undoing. There’s
a lot that is grey - the exclusive focus on large cities or on rural areas
misses out on the rural towns and the other cities. I don’t want to be
dragged-into policy for we have so, so, so many experts for that. Research
does bring out the critical role in of these both as an interface and anchor
for rural prosperity and balanced regional growth.
TNS: It seems as if this
is not a Pakistan-specific issue. The discussion everywhere around the globe
is on urbanisation. Is there an assumption that ultimately the entire world
is going to urbanise?
RA: Well, it is also said
that the cropping and livestock requirement of the world can be produced in
very small space while cities constitute not more than 1 per cent of the
land surface of the entire world. In the US, a vast space, more than 70 per
cent of its population lives in and around cities and so it is in Europe.
India is different.
Academics now insist that in ‘shining India’ a certain class glitters in
the largely urban based ‘rise’ of India. Over 80 per cent of India’s
billion and a quarter people live in settlements of 200,000 or less and they
are not in the high-skilled businesses that make India to be a dominant
economic player today. Yes no doubt there is some trickle-down to the
otherwise less fortunate but still the India of the 85 per cent is different
from the shining India of the 15 per cent who glitter. So growth is
necessary to shine, but something more seems essential for all to bask in
Our own work on spatial
geography of Pakistan shows that inequalities across space appear to be
increasing — especially when you look at rural areas or what we call
underdeveloped areas or less fortunate areas — and when these happen to be
the whole of Balochistan, the Seraiki sub province and large parts of Sindh
TNS: How much of
government intervention is involved in this urbanising phenomenon that you
are talking about?
RA: Once again, you want
to drag me into policy! From what I understand, without government, nothing
happens. To me, the way our large-scale private sector is organised, be it
manufacturing, trade or services, it feeds off the public sector. In any
case, municipal works will remain a public sector activity, so will
development of most infrastructure.
TNS: You had talked about
the lack of debate regarding inequality? And aren’t policies supposed to
be made after a lot of debate?
RA: Yes, there is debate
and there is debate. We had a debate on inter-provincial inequality and the
kind of academic work that was done then has in my view not been produced
since; but “all debate, no change” stoked the political fires that led
to Bangladesh. Policies are being made all the time: okay, forget about
policy, let’s say action is being taken all the time and we don’t wait
for consensus to develop. For instance, in reducing tax rates for the rich,
an action has been taken, without a debate; we give rebates to industries,
whatever the reason, good or bad, we don’t wait for consensus, nor do we
seek any. Academics stay engaged in debates, as they must, but shouldn’t
there be action regarding inequality? In any case, inequality is not about
policy, it’s a matter of basic principle.
When we get out of the
petty rural-urban squabble, and such others, the real, tough, question
awaits us — that of inequality — sub-national, sub-provincial, regional
— a reflection of the social, economic and political inequalities in
Pakistan. Personally I believe we are past the question posed in the title
of Tariq Ali’s book ‘Can Pakistan survive?’ In the wake of increasing
inequality between people, between households and between regions, and the
general apathy towards it, the lack of debate on it, the question that
suggests itself could well be ‘should Pakistan survive?’ Should a people
so apathetic to inequality survive — as a nation, community, family, even
For some five
decades, development planners in Pakistan looked at rural and urban areas as
if the two had to be protected from each other. Nearly all rural development
programmes, from the rural works programmes of the sixties to the
Parliamentarian’s and local bodies’ programmes of later years, stated
the containment of rural to urban migration as a key objective. The migrants
only added slums to the cities and were a source of congestion. Urban
development worried about slum clearance and decongestion.
In effect, these
programmes were an addendum to the overall strategy of development, designed
to cater to the rural and urban lobbyists. Regardless, the overall strategy
was based on industrialisation and structural change by increasing the share
of manufacturing in the national output. In 2000s, devolved local governance
did away with the administratively defined urban-rural divide. But the
divide has returned in the present decade, not only in local governance, but
also in the thinking of planners.
Limiting the ‘urban’
to cities as engines of growth, the 2011 Framework for Economic Growth
included them in its count of four pillars. More recently, there is a shift
from engines to drivers. Vision 2025, the latest arrival on the scene, has
eleven 21st Century Drivers of Growth. Cities are not among them. Where the
action lies after the 18th Amendment, the provincial planning documents
continue to talk of cities as engines of growth.
Underlying the rural-urban
dichotomy is the failure to look at urbanisation as a process. No city is an
island. Research based on the results of the last census in 1998 showed that
the urban population is more likely to be over 40 per cent of the total
instead of the 32 per cent in the census.
By far, the most important
research programme has been pioneered by Reza Ali since 1999. Taking note of
the changed definition of ‘urban’ in the last two censuses, he brought
into sharp focus the cumulative effect of the urbanisation process that was
underway. Recently, this programme has moved from understanding the
urbanisation process to estimation. Taken together, the results now provide
policy makers with a firmer basis to re-examine their ‘biases’ and
‘tilts’ to consider a realistic framework.
In the 1970s, Lipton and
Bates led the chorus about an urban bias in development policies and plans,
nurtured by pressures from organised urban groups to promote their
interests. They provided intellectual sustenance to the neo-liberal advocacy
of agricultural exports. Nolan and White (1984) pointed out in the Chinese
context the complexity of rural-urban realities and the presence of “state
What is dubbed “new”
economic geography focused on the rising spatial inter-dependence. Duranton
(2009) stressed on “broadening the focus from within city to between
cities by reducing obstacles to reallocation across cities”.
As noted earlier, an
important implication of the ‘rural bias’ has been (other than
suggesting that cities are intrinsically evil and, therefore, bad) to try to
stop the outflow of people by directing investment to rural areas. We need
to recognise the symbiotic relationship between the rural areas and cities,
and to clarify our ideas regarding development.
A society engaged in
subsistence agriculture is unlikely to develop surpluses unless there is
demand for them. It is not agricultural surpluses that sustain cities, but
the reverse. What would be the possible point of a surplus if everyone were
engaged in agriculture? As agriculture becomes more efficient, fewer workers
are required. Surplus workers must look to non-agricultural activities for
their survival. Studies show that less than 30 per cent of the future growth
of rural population can be absorbed in the agricultural sector.
There is nothing to
support the notion that cities have become too big, as if there were an
ideal size for them. All available evidence suggests that urban size and
efficiency are not mutually related. By suggesting that investing in rural
areas should have priority, development funds allocation has been
misdirected away from urban areas where the same investment, better managed,
would have produced better results for the benefit of the poor and the
A partial adoption of
Euro-centric settlement patterns has also been inimical to development,
based as these are on concepts that assume a flat, feature-less plain with
settlements achieving a hierarchy determined by spatial placing and
distance. Consequently, rural
development resources are located in “central places”, such as would
ensure equi-access to the population living around them.
concentration of population in recent years has been what Reza Ali calls
“ribbon development”. Urban areas are not merely those that fall within
given administrative boundaries, but include the “urbanising”
populations that line either side of the radial inter-city roads. To ignore
these phenomena and direct health, education, and other social investments
primarily to rural areas is ill-advised.
Being urban is not merely
a physical fact; it is also, and perhaps more importantly, a state of mind.
It is in this respect that we particularly underestimate the importance of
our urban population. Access to urban facilities, and in particular
information, is dictated by access to transport and communications.
Increasingly, this means that those living along inter-city highways are far
more urbanised than those in more “central” places.
Currently, we have been
ignoring these ribbon settlements in favour of central places, and directing
our resources accordingly. Our
national planners justify this by assuming that stopping the flow of
internal migration to the major urban areas and redirecting it to secondary
cities is somehow more desirable. The reality is that despite official
exhortations and monetary concessions, the majority of our teachers and
doctors do not want to live and work in the secondary cities.
To conclude: increased
urbanisation is not a threat; it is perhaps the only glimmer of hope that
we, as agriculture-dominated economy, have. Secondly, an inadequately
understood spatial development policy will waste scarce resources,
committing them to areas that do not correspond to the real needs, demands
and expectations of our people.
Thirdly, attributing urban
problems merely to their size and magnitude is misleading and blurs the
overall perspective for national planning, development and priorities in the
matter of allocation of resources. Fourthly, by implying that urbanisation
is somehow excessively rapid and chaotic allows urban managers to hide their
own inadequacies instead of forcing them to take a realistic approach that
is inclusive, participatory and decentralised, and far less bureaucratic,
patronising, reactive and restrictive.
Finally, we are not
arguing against the development of agriculture. Nor are we saying that
cities are necessarily the engines of growth. We are merely warning against
the simplistic notions of the ‘rural’ and ‘urban’ in matters of
The authors have served,
respectively, as Chief Economist of the Planning Commission of Pakistan and
Director of Development Planning Unit, University College, London
consensus has yet emerged about the degree of urbanisation in Pakistan,
there is a growing agreement among social scientists that the use of
administrative criterion to define the term ‘urban’ has led to an
underestimation of urbanisation. The recent article of Reza Ali,
‘Estimating Urbanisation’ has reinforced it.
administrative-based definition, ‘rural’ is always a residual category
after defining ‘urban’. Ali has not only redefined
the term ‘urban’ based on density and proximity but also has defined
‘rural’, with an identification of a new category of ‘urbanising
areas’. One can differ with him on definitions used for all three terms
— ‘rural’, ‘urban’ and ‘urbanising’, but, his thinking is very
close to the ground realities.
In 2003, I used both the
micro-data files of the 1998 Population Census and the published reports of
the earlier censuses to evaluate the use of administrative criterion for
defining ‘urban’. The first conclusion was that the strict use of
size-specific criterion, giving all rural localities having 5,000 or more
persons the status of urban, as claimed in first three population censuses
(1951, 1961 and 1972), has never been applied. In fact, in the 1951 to 1972
censuses only a small number of localities outside the notified urban
localities was declared ‘urban’.
If the populations of
these small numbers of localities are excluded from the total urban
population of the country it makes no real difference on the level of
urbanisation, which, after adjustment, declined only marginally. So,
practically, in all five population censuses the administrative criterion
has been used to define ‘urban’.
The use of size-specific
criterion in the 1998 Population Census, giving all rural localities having
5,000 or more persons the status of urban, would almost double the urban
population. But, it can be argued that in terms of other characteristics,
these localities are predominantly rural.
I tested this argument by
comparing the characteristics of the 210 declared ‘urban’ localities,
which inhabited 20,000 or fewer people with those rural localities that were
inhabited by 5,000 or more persons and were not given urban status in the
The analysis found that
361 rural localities were better than many declared ‘urban’ localities
in terms of the urban-related characteristics. A close look at these rural
localities shows that some of them were adjacent to a large urban centre.
The treatment of these localities as rural cannot be justified; it leads to
underestimation of the overall level of urbanisation in the country, as well
as the actual population of the large cities.
Some other large rural
localities having 20,000 or more persons are, in fact, trade and industrial
centres situated at the Grand Trunk (GT) Road; in all aspects these are
towns. Some rural localities, not located even on GT Road, are quite large
towns having more than 50,000 people.
The analysis further found
that the 361 rural localities inhabited by 5,000 or more persons having
urban-related characteristics are concentrated in four districts of central
Punjab — Lahore, Faisalabad, Sheikhupura and Gujranwala. Few localities
are also found in Rawalpindi and Attock districts in northern Punjab. In KP,
the concentration was found in Peshawar, Mardan, Charsadda, Nowshera, and
Malakand districts. In Sindh and Balochistan, the districts of concentration
were Hyderabad and Quetta respectively.
Overall, the rural
localities that have urban characteristics are located in relatively more
developed districts of the country. It appears that Reza Ali has
incorporated this geographical concentration of urban-type rural communities
is his concept of ‘urbanising areas’.
What would have been the
level of urbanisation if those 361 rural localities that were better than
many small urban centres in terms of urban-related characteristic were
treated as urban in the 1998 Population Census? The overall share of urban
population would have increased from the recorded level of 32.5 per cent,
based on the administrative criterion, to 36.2 per cent.
I further estimated that,
if the excluded rural communities having urban characteristics were treated
as urban, and rural localities adjacent to large urban centres were
considered as part of these centres, the total urban population in 1998
would be around 51 million.
The share of urban
population in the total population would approximately be 39 per cent in
1998, with largest increase in KPK and lowest in Sindh. If this figure of 39
per cent is considered as base for the degree of urbanisation in 1998 and
urban growth rate, which is presently much higher than the rural growth
rate, is applied to project urban population, the degree of urbanisation in
Pakistan today would be much higher than the level projected in official or
What Reza Ali has proposed
in his article for defining the terms ‘rural’, ‘urban’ or
‘urbanising areas’ may not be a perfect solution for estimating
urbanisation in Pakistan, because its application alters markedly the
province-level estimates. However, he has given some practical suggestions
for the estimation of the degree of urbanisation.
The writer is Joint
Director, Pakistan Institute of Development Economics