# 36382Empty Colonies Don’t Shop: How India’s Land Conversion Is Quietly Redrawing the Retail Map
India is pricing land for a future that arrives slowly — and retail is paying rent on that future today.
India is converting farmland around its cities into plots faster than people move in. Land under non-agricultural use has tripled since 1950-51. Chitrangana’s research, models and probability-rated outlook on what vacant colonies, sprawl and vertical density mean for retail site selection.
not a sales representative.
For retailers, franchise operators, commercial developers and anyone planning a new store on the edge of a growing city. Around India’s tier-2 and tier-3 cities, farmland is turning into plotted colonies and shop complexes faster than people are moving in. For a retail business, that gap matters more than any land-price chart. A plot is not a customer. An occupied home is.
India is converting agricultural land around its cities into residential and commercial plots well ahead of real demand. Land under non-agricultural use has tripled since 1950-51 to about 28.5 million hectares, while net sown area has not grown in fifty years. Many of these colonies stay largely empty for years, because land is being bought as a store of value, the way gold is. Chitrangana’s read is that this quietly breaks retail expansion: shops open where the boards say “upcoming”, not where people actually live, and horizontal sprawl spreads demand too thin for most stores to survive. Over the next decade, we expect retail site selection to move from land price to occupancy, and compact, vertical, mixed-use pockets near transport to become the strongest retail locations in tier-2 India. Density helps only when real demand exists, so occupancy, not floor space, is the number to watch.
The Pattern Every Growing City Now Shows
Drive 10 to 15 kilometres out of almost any expanding tier-2 city in India. You will see the same picture. Boundary walls, entry gates, streetlights, paved internal roads, and a hoarding that says “possession ready”. Then you notice how few houses are actually lived in.
This is not a one-city story. In our work across Indian tier-2 and tier-3 markets, we keep seeing new colonies on the outskirts that remain mostly vacant for many years, often close to a decade. The land was farmland a few seasons ago. It stopped producing crops the day it was converted. It has not yet started producing a neighbourhood.
The reason is simple. A large share of these plots are bought to hold, not to live in. Land in India behaves like gold: a safe place to park savings, and sometimes unaccounted money, that is expected to rise in value. Nobody needs to build on it for the investment to work. So nobody does.
Census 2011 data, as analysed by the Centre for Social and Economic Progress (CSEP). The next Census will show how far the gap has moved; field observation in peri-urban tier-2 belts suggests it has not closed.
The cost does not stop at empty plots. When farmland around a city is converted, groundwater recharge falls, nearby water stress rises, and the farmer who has just received a high price has little reason to keep investing in his remaining crops. The land is lost to agriculture years before it is gained by the city. In between, it produces nothing for anyone: not food, not homes, and not customers.
Seventy Years of Data: Where India’s Land Has Gone
The field observation is backed by India’s own land records. The Directorate of Economics and Statistics in the Ministry of Agriculture has tracked how every hectare of the country is used since 1950-51. One line in that record tells the story clearly: land put to non-agricultural uses, meaning buildings, roads, railways, industry and settlements.
Land Use Statistics, Directorate of Economics and Statistics, Ministry of Agriculture & Farmers Welfare (2023-24 provisional). Projections: Chitrangana trend model at the recent and the 2000s pace of conversion.
Land under non-agricultural use has tripled since independence, from about 9.4 million hectares to about 28.5 million hectares. The pace has not slowed much. Between 2010-11 and 2023-24 alone, about 2.1 million hectares more went under construction and infrastructure. That is roughly 21,000 square kilometres, more than five times the area of Indore district.
At the same time, the land that actually grows food has stopped growing. India’s net sown area reached about 141 million hectares in the early 1970s and has stayed close to that level for fifty years. The latest provisional figure shows a dip.
Land Use Statistics 2023-24, Directorate of Economics and Statistics, Ministry of Agriculture & Farmers Welfare. The 2023-24 figure is provisional and partly reflects higher fallow land that year, so it should be read as a warning sign, not yet a trend.
So the city is not growing into empty land. It is growing into farmland, and India has very little room left to add new farmland anywhere else.
Urban growth is real, but slower and messier than the plots suggest
India is urbanising, but at a steady pace, not an explosive one. The share of Indians living in urban areas rose by about 14 percentage points in sixty years of Census data.
Census of India, 1951–2011. Shaded range: Chitrangana projection from linear and exponential trends fitted to Census 1981–2011.
Two things matter for business here. First, the number of people moving into cities each year is far smaller than the number of plots being created around them, which is why so many colonies wait. Second, a lot of urbanisation is happening without a plan. The number of “census towns”, places that have become urban in character but are still governed as villages, almost tripled in one decade.
Census of India. Census towns are settlements with urban population size, density and non-farm work, but without an urban local body. They grow without the planning, zoning or infrastructure that a city gets.
- Land under non-agricultural use: projected from the 2023-24 level of 28.5 Mha at two observed historical speeds: the recent pace (about 0.16 Mha a year, 2010-11 to 2023-24) and the faster 2000s pace (about 0.27 Mha a year, 2000-01 to 2010-11). Result: 30.5–31.8 Mha by 2036 and 32.2–34.7 Mha by 2047.
- Urban share: a linear and an exponential trend fitted to Census 1981–2011. Result: 37–39% urban by 2036 and 40–44% by 2047.
- What it means: another 2 to 6 million hectares, mostly farmland around cities, is likely to change use by 2047. Whether it becomes productive neighbourhoods or vacant colonies depends on how expansion is sequenced.
- Limits: trend models assume past speed continues. Policy changes, a new Census count, or an economic shock could move these numbers either way. They are planning ranges, not forecasts.
Why This Is a Retail Problem, Not Only a Planning Problem
Most people read this as an urban-planning issue. It is also one of the most expensive mistakes a retail business can make, and it happens in three ways.
1. Retail follows the hoarding, not the household
A new colony launches, a commercial complex comes up at its gate, and shop owners, franchisees and brands buy or lease space because the area is “developing”. But a shop needs footfall every day, not someday. When occupancy takes eight or ten years to arrive, the shop either closes, survives on losses, or sits shuttered. The investor in the shop pays for the speculation of the investor in the plot.
2. Horizontal sprawl spreads demand too thin
Take one lakh people. Spread them across 40 square kilometres of plotted colonies and they support far fewer viable shops than the same one lakh people living within 10 square kilometres. In the sprawled version, each store needs a bigger catchment, walk-in traffic is lower, delivery costs more per order, and the neighbourhood never reaches the density where a pharmacy, a grocery, a salon, a café and a clinic can all survive on the same street.
3. Vertical, planned growth concentrates demand
When a tier-2 city grows upwards in planned pockets instead of outwards in scattered plots, more households sit within walking distance of the same street. That is what makes a local retail ecosystem possible. More customers per square kilometre means more shops can earn a living, and more small businesses can exist. This is the economic argument for vertical expansion in tier-2 and tier-3 cities: not taller buildings for their own sake, but enough people in one place to keep a market alive.
Every vacant colony on the edge of a city is a retail catchment that exists on paper and not on the street. Businesses keep reading land prices as demand. Land price only tells you someone expects demand. Occupancy tells you it has arrived.
Nitin Lodha, Principal Business Architect & Strategist, Chitrangana
Modelling the Wait: How Long Does a Shop Wait for Its Customers?
To make this practical, we built a simple model. Take two developments with the same 2,000 homes. One is a plotted township spread over a wide area on the edge of the city, where most buyers are holding land. The other is a compact, vertical, mixed-use pocket closer to existing roads and transport, where most buyers move in. Now ask one question: when does a daily-needs store at the gate become viable?
Illustrative model with stated assumptions; see the assumptions box below.
- Both developments have 2,000 homes. A daily-needs store (grocery, dairy, pharmacy) is assumed to need about 600 occupied homes within its catchment to cover rent, staff and stock.
- Sprawled township: occupancy builds slowly and levels off around 80%, typical of plots bought to hold.
- Compact pocket: occupancy builds quickly and levels off around 90%, typical of homes bought to live in.
- Result: the compact pocket crosses the viability line in about 2 years. The sprawled township takes about 8. A shop opened on day one in the township pays roughly six extra years of rent before it can stand on its own.
- This is an illustrative model with stated assumptions, not measured data. Use it to test your own location by replacing the numbers with local occupancy counts.
The Honest Limit: Density Without Demand Is Just Taller Vacancy
Vertical growth is not a magic answer, and it is important to say so clearly. If a new complex adds 1,000 shops to an area that has spending power for 400, the other 600 do not become businesses. They become empty shutters on a higher floor.
An illustrative example, not measured data. Floor space does not create spending power; occupied households and local incomes do.
So the real variable is not height, and not land. It is effective demand: how many people actually live, work and spend in a catchment today. This is also why our wider thinking on this subject argues for development permissions that open in phases, linked to how much of the already-approved area is occupied and serviced, rather than letting every ring around a city open at once. A city that fills one zone before opening the next produces fewer empty colonies and, for retail, fewer dead streets.
A Simple Test Before You Open a Store on the Edge of a City
For any retailer, franchisee or entrepreneur evaluating a peri-urban location, these checks matter more than the brochure or the land rate.
| Check | What to Look For | Why It Matters |
|---|---|---|
| Lights at night | Share of homes actually lit after 8 pm, counted on two or three different evenings | The fastest, cheapest occupancy count there is |
| Utility connections | Active electricity and water connections versus plots sold | Separates residents from owners |
| Daily-need shops already open | Grocery, milk, pharmacy, vegetable vendors trading for 12+ months | Proof that daily spending exists |
| School and transport | Nearby school admissions, bus or auto routes in daily use | Families settle where these exist |
| Absorption pace | How many new homes became occupied in the last 12 months | Shows whether the catchment is growing or waiting |
If a location fails three of these five, it is a land investment, not a retail location. That can still be a reasonable decision for an investor. It is rarely a good one for a shop that has to pay rent and salaries every month.
What We Expect to Happen Next — Probability and Prediction
These are Chitrangana’s assessments, based on the pattern above, our field work in tier-2 and tier-3 markets, and the direction of existing policy such as India’s National Transit-Oriented Development Policy (2017), which already names horizontal sprawl as a problem. They are judgements, not official forecasts.
| Prediction | Probability | Horizon | What It Means for Retail |
|---|---|---|---|
| Occupancy data (utility, delivery and mobility signals) becomes a standard input in retail site selection | High | 2–4 years | Brands and franchise networks that measure real occupancy will open fewer failed stores |
| Outskirt commercial complexes stay oversupplied, with soft rents and long vacancy | High | Ongoing, 5+ years | Tenants gain negotiating power; investors in shops on the edge face long waits |
| Compact, mixed-use, vertical pockets near transport become the strongest retail addresses in tier-2 cities | Medium–High | 5–8 years | Smaller, denser catchments beat large plotted townships for footfall |
| Quick commerce and dark stores expand in tier-2 only where density crosses a viable threshold | High | 2–5 years | Sprawled colonies will be served late, or not at all, by fast delivery |
| Some states begin linking new development permissions to occupancy and infrastructure readiness | Medium | 5–10 years | Fewer dead colonies; more predictable retail catchments in newly opened zones |
| Another 2–6 million hectares of mostly peri-urban land changes to non-agricultural use by 2047 (Chitrangana model) | High | By 2047 | The retail map of every tier-2 city will be redrawn; the question is where people actually settle within it |
| India crosses 37–39% urban by 2036 and 40–44% by 2047 (Chitrangana model) | Medium–High | 2036 / 2047 | Urban demand grows steadily, not explosively; supply built faster than this will stay empty longer |
What This Means for Business Owners and New Entrants
If you are expanding a retail business, starting a franchise, or moving from a traditional trade into a new format, three practical shifts follow from this.
- Buy into people, not plots. Choose locations by how many households live within a 1–2 km walk or ride today, not by how many plots have been sold.
- Prefer density over distance. A smaller shop in a dense, occupied pocket will usually outperform a large showroom on the empty edge, even at a higher rent.
- Phase your own expansion the way cities should. Open the next store only when the current catchment proves its footfall. The discipline that would fix city sprawl also fixes retail sprawl.
The deeper point is simple. India’s land is being priced for a future that arrives slowly, and many retail businesses are paying rent on that future today. The businesses that win the next decade in tier-2 and tier-3 India will be the ones that learn to read occupancy, not optimism.
Frequently asked
Why are so many colonies on the outskirts of Indian cities empty?
How does urban sprawl affect retail businesses?
Is vertical development always better for retail?
How much land in India has moved out of agriculture?
How should a retailer choose a location in a new peri-urban area?
Wondering where your business sits in the commerce shift?
We map how ready you are today — and design the architecture that keeps you the answer, not the afterthought.
Explore our consulting
