Thursday, March 24, 2011

Revamping urban housing policies

A McKinsey Report (summary here) in 2010 estimated that 25 million urban households cannot afford housing, and estimates the demand to rise to 38 million by 2030. The Government of India have estimated an additional housing requirement of 26.53 million during the 11th Five Year Plan (2007-12).

Bridging this deficit is one of the biggest challenges facing urban administrators and infrastructure specialists in the country. Further, in view of the growing importance of urban growth (urban areas contributed 58% of GDP in 2008 itself), this is critical to sustaining our economic growth itself.

State governments across the country have adopted several policy variations to address this. The commonest strategy has been direct construction of housing units, either on vacant lands or on existing slums (by in-situ development). State governments have dovetailed funds from various Central Government schemes and bank loans to develop large numbers of housing units.

Another increasingly important approach has been to develop slums on public-private-partnerships (PPPs). The most popular form is one where developers are permitted to commercially develop a portion of government land allotted to them in return for constructing a defined number of dwelling units for the poor in the remaining land.

Other strategies include earmarking a portion of a layout or built-up area being developed for urban poor. The private developers would then sell these houses to the targeted category at prevailing market rates. It is hoped that this would both open up scarce urban land for housing and also cross-subsidize weaker section housing. As part of efforts to acheive the goal of "Affordable Housing for All", the National Urban Housing and Habitat Policy, 2007 (NUH&HP) mandates the reservation of "10-15 percent land in new public/ private housing projects or 20-25 percent of FAR (whichever is greater) for EWS/ LIG housing through appropriate legal stipulations and special initiatives".

The Andhra Pradesh Government has become only the latest state to notify guidelines making it mandatory to reserve 20% of developed land in all urban housing projects to the economically weaker sections (EWS) and Lower Income Groups (LIGs). This is in addition to 15% of built-up area being reserved for EWS and LIG in housing projects.

For the layouts, the maximum plot size for EWS is to be 35 sq.mts and 55 sq.mts for LIG. The plots are to be disposed to registered weaker section societies or to public agencies at prevailing market rates as per the Registration Department. Such societies should develop them as group housing schemes and not plotted development. The policy also provides for complete exemption on stamp duty, non-agriculture conversion charges, and development charges for one time registration of EWS/LIG and 50% of development charges and other fees for LIG plots.

I am afraid that such policies, while populist and logically appealing at first view, do not pass muster on rigorous analysis. For a start, they are cosmetic exercises and will do little to address the urban housing problem. Even if all the several formidable implementation problems are overlooked, the houses that can be built under this will be minuscule in comparison to the requirements, especially in the larger cities.

The fundamental objective of all these interventions should be to ensure that it increases the supply of urban housing stock with the least possible market distortions. However, such policies will not only fail to meet its objective, but will also generates market inefficiencies that can create other problems.

There are two categories of consumers who face massive urban housing shortages. The economically weaker sections, consisting mainly of migrant labor, exert the largest demand on the market. The lower-income and middle-income groups too form large shares of the demand for urban housing.

In fact, such strategies overlook the multiplicity of categories within urban poor themselves. At one end are those, primarily the LIG, who can possibly afford to buy these houses with a bank loan. At the other end of the spectrum are the largest numbers, especially the poorest, who cannot afford these houses and will be reliant on heavily subsidized housing.

More importantly, this will adversely affect the general housing market itself and conflict with the primary objective of making housing affordable for the non-poor. From the perspective of the private developer, earmarked development will effectively reduce the amount of land available for full realization of commercial opportunities.

The developers would naturally pass on this additional cost (by way of incomes foregone, the opportunity cost) to purchasers through higher prices. In other words, the buyers of the original housing units (who are not always the economically well-off) end up paying higher prices. And this too without the LIG and EWS necessarily getting houses at a low price (even if they get at low price, the government pays a very high subsidy).

In other words, the earmarking serves as a tax on home buyers. The middle-income group (MIG), themselves a large customer group and the engine that drives urban economic growth, will be the worst affected by this. Similar attempts to get developers part with a share of their housing, say by offering a share of developed space for social housing in return for additional FAR also translates into higher prices for buyers of the original units.

In any case, the extent of land that is likely to be made available through such interventions is too small to impose such a large cost on the general market itself. Finally, there is also the administrative nightmare of enforcing such programs. The wide variations in land values, itself non-transparent, complicates matters. Who are the LIG and EWS? How do we monitor the allocation process? Even if state governments agree to provide the subsidy, how will the price discovery happen and what will be subsidy? How do we ensure that these housing units are not captured by middlemen who in turn will rent it out to beneficiaries?

At a more fundamental level, the massive demand can be met only with a mix of policies that increase the supply of vacant lands and encouragess vertical growth. Since most cities have already run out of their stock of vacant and un-allocated government lands, it is important to tap the massive extents of un-utilized lands available with various government departments.

Another source for unlocking land is to re-develop the squatter settlements on government lands, both notified and un-notified slums, with multi-storied housing units. It is here that we run into the highly regressive regulatory restrictions on built-up area that can be developed in any land.

There are regulatory and infrastructural limits on vertical growth in our cities. Indian cities have amongst the lowest Floor Area Ratios (FAR), ranging from 1-3. In contrast, FSI in most Asian cities varies from 5 to 15 and in many Western cities goes up to even 25. However, such vertical expansion would also require investments in the appropriate enabling utility infrastructure.

Restrictive Floor Area Ratios (FAR) are the single biggest impediment to unlocking the potential of urban housing market in India. Among all the bigger economies, India has the lowest FAR, which restricts the amount of built-up area that can be constructed on a land. The commonplace manifestation of this is the absence of high-rise buildings in our cities. I have blogged earlier about the need for Indian cities to go vertical and infrastructure facilities to be planned to accommdate vertical growth.

As a recent World Bank paper points out, all these policies are a legacy of an era when the objective was to discourage urban migration and distribute growth to smaller cities and villages.

There are also other stifling policy restrictions. The McKinsey report found that state and central governments impose a total tax of 27% on housing. This is in addition to the several regulatory conditions that are also de-facto taxes that increase the real cost of urban housing.

Update 1 (30/6/2012)

The Andhra Pradesh government has modified its earmarking requirements government order by replacing it with some form of fee equivalent. Now, if the project area is over five acres, the builder has to provide 10 per cent of the total built-up area for the economically weaker section or they can reserve some units. Further, developers taking up projects above 3,000 sq.m. and up to five acres have to pay shelter fee to the respective municipal bodies. The builders have to pay a shelter fee of Rs 400 per sq.mt metre up to 750 per sq.mt depending upon the project location.