How is the TAPI pipeline project
The gas sale purchase agreements (GSPAs) have been signed. What they are doing now is to form a consortium for the pipeline and get a lead partner. It is contemplated that the pipeline will be run run by an international company, and not any member country. Each country will have a minority stake in the pipeline. It is contemplated that they will probably have composite of 40-49% stake, while 51-60% stake will be of the contractor/pipeline developer.
Is the consortium company going to be registered in the US
Likely not in the US. There is this treaty for protection of pipelines and certain countries are signatories to that. The US is one of that, but the consortium may be registered in an island country.
What is the transit fee model for gas flowing through the pipeline
It will be a per unit of gas (mBtu) transit fee. Within India there wont be any transit fee because the TAPI pipeline technically will end at the borders of India and then it will be Gail pipeline. So to Gail it will get delivered at the border. India will have to pay the transit fee to Afghanistan and Pakistan.
Will IOC partner with Gail to sell
There are some conversations on that. But lets see how this story develops. IOC wants to get into gas business. It is trying to enter into LNG sale-purchase and to do some project in North America. It is booking capacity and trying to develop an LNG regasification terminal in the east coast. Since both Gail and IOC are government companies and it depends on what the ministry thinks and to what extent it allows them to do that.
When do you expect work to begin on the pipeline
It all depends on the interest shown by stakeholders, and who is the transporter. The transporter will have to to manage the relations with Turkmenistan, Afghanistan and Pakistan. Not so much with India as delivery is at its borders.
Do you think the deadline of 2017 given by the previous petroleum minister for flowing gas via the TAPI line would be a reality
That may be difficult. We are already in 2014 and there is significant ground to be covered. Lets assume that it takes four-six months to identify the pipeline contractor and tie up the contracts and start. The pipeline would then take close to three years for completion. So maybe the project will get pushed by six months to one year from 2007, depending on when the construction begins.
Infrastructure projects have suffered in the past few years. What kind of infrastructure push you expect with the new government taking charge
Infrastructure is a key issue in the BJP manifesto and development does seem to be high on its agenda. Gujarat has done relatively better in terms of infrastructure. Ports, regasification terminals, pipeline networksa lot has been done in the state. But, infrastructure is a sector that cant get a kick-start overnight. It is bit of a gradual process. Each award of any kind of contract takes time. The lenders are looking upbeat towards investment in the infrastructure sector.
The sector has its share of problems. If you have to kind of prioritise, which area would need an immediate course correction
If you look at infrastructure projects, some have kicked off, while a few have run in to problems and delays. The critical part about infrastructure projects is that these are long duration projects, which by the very nature cant be very rigid in terms of contract. Irony is, every two-three years, there is either a recession happening or growth getting hampered or financing getting affected or a sudden policy change. This is not just in India but worldwide. The contracts need to have a degree of flexibility in risk allocation. It is very difficult to predict a contract, which is going to remain static for the next 20 years as the market doesnt remain static for 20 years. In some areas, where the regulator has been doing well, those sectors have seen to have performed well. Because at the end of the day, the job of the regulator is to make sure that any act of regulation is done basis the given circumstance/time and vary its decisions depending on the time. This whole concept of model contracts and risk allocation being very strict is not feasible. Because if the market changes, those risks become larger than anticipated by the developer and is beyond his capacity. Then he either wants to re-negotiate or not going to perform or lenders are not going to lend.
Would you say that our contracts are not mature enough or not dynamic enough to factor in changes over time
Over a period, we have been trying to do a contract which is a model for everything for each sector. If you look across, these model contracts are principally based on the model contract of a particular sector. To my mind, an airport concession is very different from a road concession, which is very different from a power purchase agreement or a port concession. Even in those, one port is different from another port. One airport is different from another airport. The whole concept of a model concession in which you have to make minimal changes and the idea of one fit works for all is a bit of a problem. A contract is nothing but sharing of risk between the parties. Risk allocation needs to be determined on a project basis. If each project is unique then the risk allocation of each has got to be unique. Yes, there will be broad similarities in the same sector but still each project is a different project.
You have worked on airport transactions. How has modernisation of airports succeeded
In the airport sector, though the traffic risk is taken somewhat but there is a regulator in-place. The regulator (Airport Economic Regulatory Authority) will look at the traffic and set a particular charge so that the developer can recover the investment. So you are making a forward looking statement on traffic. AERA would not monitor traffic but it would look at the historic traffic charts and then come out with a suggestive pricing. In short, you are not taking the traffic risk for the next 20 years today, whereas in the roads model, you are taking traffic risk on the day you present your bid. And if the traffic doesnt come in then it won't do well. Suddenly those projects will become completely non-viable because they are viable only on a particular set of data, and if the data is wrong the project becomes non-viable. So, that has been the basic difference between the two sectors.
What are the things that need to be addressed before auctioning a project
The government should do the groundwork, as it reduces the risk. Land acquisition, environmental clearances can be taken upfront. The project can be put into a special purpose vehicle and then bid out.
Would you prefer long-term infrastructures to be reviewed every five years
It is a good idea, if the mechanism ensures a fair review and adjustments are made to match the change in situations.