On the same day that the finance minister presented the Union Budget for 2005-06, the Reserve Bank of India [Get Quote] announced a roadmap for foreign banks to expand their presence in India.

As many people expected, it is an extremely cautious and guarded opening up until 2009, with some assurance of widening the opportunities after that.

The main elements of the first phase are with regard to organisational structure, allowing foreign banks to transit from their current branch framework to the wholly owned subsidiary model.

During this period, they will be allowed to acquire ownership and control in banks that are considered to be in need of "restructuring" by the RBI. In such banks, the RBI would work out a plan for increasing the foreign bank's ownership stake up to a maximum of 74 per cent.

From 2009 onwards, assuming that this roadmap is still in effect, the freedom to acquire up to 74 per cent would be extended to any private bank.

The reason for this cautiousness is the perception that the Indian banking sector will need the time to prepare itself for the competition that will ensue once entry by foreign banks is fully opened up.

As with many reform measures, there is a strong case for gradualism as a way to minimise the costs of adjustment. But, in this situation, as with all in which gradualism is essentially a compromise with incumbent interests, there has to be a strong element of reciprocity.

If one accepts the premise that competition is good for consumers, delay in introducing it clearly hurts consumer interests. Domestic banks, which have been given this four-year breathing space, cannot just sit back and wait for the change to happen.

They have an obligation to consumers of banking services to use it to restructure themselves and achieve levels of efficiency that will make them strong players in the post-2009 scenario.

Not that Indian banks, particularly the public sector ones, haven't been making attempts to do so. Some have aggressively ventured into retail finance, the hot growth segment over the last few years, but one that required strategies and skills not usually associated with the public sector.

All of them participated in the voluntary retirement scheme in 2001, which saw a reduction of over 10 per cent in the aggregate workforce.

A few days before the roadmap was announced, the scope of autonomy of bank managements was expanded to allow for limited exit from branches, expansion abroad, and competing for talent in the marketplace without being constrained by the banks' pay scales.

However, as welcome as these moves are, they are way too slow and cautious. Gradualism towards foreign investment may be justifiable, but only if the process of domestic adjustment is rapid.

If Indian banks are slow in making the necessary transition, 2009 could be brutally destructive of them; worse, their unpreparedness could become an excuse to abandon the roadmap altogether, to the detriment of consumers of banking services in the country.

What would make for a successful transition? Do we need another roadmap for this, or will it be best to just leave it to the players themselves with the hope that the sector will converge, through trial and error, to a competitive structure four years from now?

In my view, a set of organisations emerging from decades of government control would take too long to converge to a "desirable" equilibrium. The timeframe simply does not allow for the process to work without a roadmap.

The government was instrumental in getting public sector banks to where they are today; it must also play a significant role in taking them to the threshold of 2009.

What is the optimal structure of the banking industry, if indeed there is one? We can certainly identify some key factors that would influence it. It would obviously have to take full advantage of technology, which typically comes with high fixed costs, to reduce its variable costs to a minimum.

And, it would obviously have to manage the risks inherent in its portfolio of assets, whatever this may consist of, at the lowest possible cost. Different structures may provide the most efficient solutions under different circumstances, but in today's environment, it appears that co-existence between two strategic principles fits the bill.

One, clearly, is scale. Increasing volumes by straddling every segment of banking services and even beyond to take full advantage of the investment in technology is essential for banks to become competitive.

Obviously, all banks cannot grow to efficient scales. The transition to 2009, therefore, must inevitably involve mergers between banks operating at sub-optimal scales or acquisition of small banks by larger ones. The question is: is there a reasonable consensus on what constitutes the "minimum efficient scale" in the contemporary Indian context? 

If there is, it should be relatively simple to work out various permutations and combinations between banks that would satisfy this condition and then look for maximum possible synergies.

If there isn't, and this is more likely the case, then arriving at it is the first requirement for laying out a transition roadmap. This is a complicated enough task in itself, but it is further confounded by the second strategic principle that makes for successful banking today.

Niche banking may go against the compulsion towards larger and larger volumes, but it does provide an ability to manage risk more efficiently by virtue of superior information about the marketplace for loans -- be it a region or a set of sectors.

Smaller banks may not be able to extract the same juice out of technology as larger players, but if they can offset this with lower costs of managing risks, they can survive and compete. There are examples of relatively small banks in India being quite successful.

No transition roadmap should foreclose this option. Unfortunately, for every successful niche strategy, there are several that haven't worked. Pinning down a set of viable options in the roadmap is an impossible task.

Taking all of these factors into consideration, the transition process should begin by arriving at a benchmark for an efficient scale of operations.

It should then facilitate a process of restructuring whereby banks that want to grow to that scale can explore options to merge with other banks, or acquire part of the assets of banks that might want to go the niche way.

Banks opting for a niche strategy should be allowed to go their own way and fend for themselves in the post-2009 scenario. The next four years will seem like the blink of an eye to Indian banking.