Monday, April 26, 2010

Status based taxation in local government taxes

Mark Thoma points to an excellent interview of Raj Chetty by Romesh Vaitilingam, where the economist talks about the optimal design of tax policies and public expenditure programmes.

He argues in favor of maintaining and even expanding the duration of unemployment insurance benefits for unemployed in the US. His studies on unemployment benefits indicates that the moral hazard effect (where generous European-style indefinite unemployment benefit schemes disincentivize people from looking for jobs) is offset by the "liquidity effect" (poor people have less savings and therefore being liquidity constrained they are forced into accepting any job that comes their way, whereas with some limited benefits they may be able to take some time to search out jobs more efficiently). Ivan Werning too finds much the same evidence and does not favor stringent limits on duration of benefits on the grounds that those with the longest unemployment spells are those with largest losses from foregone earnings.

On taxation policy he finds fairly strong and conclusive evidence that salience of tax increases demand elasticity. This also means that contrary to the claims that prices and tax incidence are indifferent on whom (seller or buyer) the tax is imposed, there is strong evidence that imposing the tax on buyers may minimize the impact on demand. In other words, taxes on consumers mostly gets passed through without making much dent on demand.

He also makes an interesting proposal about status-based taxation, to leverage the premiums attached to individual status in traditional societies and maximize tax collections. Prof Chetty therefore proposes offering the attraction of naming public assets to incentivize tax compliance, especially with local government taxes, in developing countries like India. However, I am not convinced about this for the following reasons

1. The local property tax rates are very small (a maximum of 1% of capital value or 20% of annual rental value of the house, both of which are in turn heavily under-valued). In fact, collectively, it is estimated that property taxes form just 0.2% of GDP. This means that there are not likely to be any large enough tax assessees even in the bigger Panchayats. Even among the handful of large houses/shops, their tax revenues are not likely to be large enough to cover any substantial asset or even a part of those assets (say a room in a school building), and this even if we consider their tax assessment for a number of years.

In the circumstances, if the panchayat decides to allocate naming rights for an asset only for covering a share of the cost of construction, it would be tantamount to selling off the few critical community assets in the village very cheap. Like divesting public enterprises at below their valuations in a depressed equity market! Or selling off at the lower-end of the value chain! The local government risks facing the considerable "seller’s curse" of having sold out a part of its own identity permanently for a far too low price in an under-developed market.

2. There are not too many community assets (both existing and proposed ones) in any village that can be so named. School building, PHC building, and community hall are the most attractive assets for naming. Street roads too can be taken up for naming, especially those where the assessee resides. However, the cost of paving a street (about Rs 20 lakhs for a 100 m street) with cement concrete is too large to be covered with property tax. A small single room school building will cost Rs 3-4 lakhs. The largest property tax assessment in a village will not be more than a few thousands, rarely more than a lakh.

3. Typically, even with small efforts, the tax collection efficiency of local bodies can be improved dramatically. Presently they neither have the capacity (most gram panchayats have just one bill collector to assess, maintain records, collect, and follow up with violators) nor incentives (property tax collections, atleast in case of small Panchayats, are a very small share of their inflows, and salaries and major works are financed with government grants) to either collect or enforce the existing tax rules. So simple efforts at tax collection and capacity building through personnel and maintenance of records will yield substantial results. The low hanging fruits are too many to be plucked.

An excellent comparison can be made with the dramatic improvements in tax governance of cities in India over the last two decades as collection and enforcement machinery have been strengthened and records updated. Further, unlike the Panchayats, salaries, repairs and maintenance expenditures and most of the infrastructure spending of cities has to come from their internal funds.

4. There is also the problem of recidivism as the assessees can slip back into their old habits and evade/avoid payment of taxes once the asset is named.

5. In any case, the competing needs of panchayats are numerous and resources too scarce for the program to be meaningfully implemented. The status tax proposal can take off only if the asset that the assessee wants to name is among that covered in the Panchayat’s priority. This coincidence of interests is likely to be very rare. The typical Panchayats priorities (which in turn is decided by the amount available), among capital assets, are most likely to be street roads and drains. Schools and hospitals are generally covered by the grants under various other schemes.

A more efficient way to capture the willingness to pay among the large assessees would be to have a formal policy that solicits donations from people in return for naming these assets. This would enable a more efficient price discovery, especially for the more prized assets (like schools, community halls, bus shelters etc), by formal allocation of the naming rights to those who bid the highest. A floor price, amounting to the actual cost (or say 75%) of construction of the building, can be fixed to formulate the rules of the bidding process. This would be a tax-plus scenario! I am inclined to believe that this policy is likely to have a number of takers, especially in small villages, where the premium attached to such symbols is substantial.

In the final analysis, the problem with local government (both panchayats and municipalities) taxation in India is that apart from the meagre property taxes and assigned revenues from stamp duty and entertainment taxes, they do not get any share of the major tax sources – commercial taxes, excise taxes, income tax and corporate taxes. Further, the local government tax rates are too small that even if they actually manage to collect their taxes, it would hardly enough to make a serious dent on their huge needs.

Update 1 (22/6/2010)
Nice post by Nancy Folbre on the debate surrounding incentives and unemployment benefits.

No comments: