Structural problems in our electricity system

Friday, 1 April 2016 00:00 -     - {{hitsCtrl.values.hits}}



The Government has appointed multiple committees to investigate the three nationwide blackouts that occurred within six months. It is hoped that they will be open to the possibility that the technical issues, such as whether a specific piece of equipment was properly maintained or not or whether adequate reactive power was available, are manifestations of a deeper malaise.

In the field of economics of infrastructure, one’s starting hypotheses when analysing problems of quality of service are that incentives are misaligned or that investment is inadequate. Having worked on regulatory reforms in electricity in 2002-04, I had some additional, even if not fully up-to-date, knowledge. 

I had asked an engineering consultant who was working on a different assignment to give me his views on the System Control Centre (SCC) in Dematagoda, having heard that it was inadequate to serve as the nerve centre of a modern electricity network. Given the need to balance fluctuating demand with supply more or less instantaneously, a SCC has to have extremely sophisticated communication capabilities. Nowadays, SCCs also make extensive use of algorithms and are significantly automated. If the SCC fails to balance demand and supply, the system will go outside the safe bounds. The result will be a blackout. 

Being a polite professional, the consultant who visited the SCC told me that we should be proud of the engineers working there under very difficult conditions.



The symptom

That was back in in 2003. When the lights went out for the third time on March 13th, I looked to see if the SCC had been modernised. Transmission reports were available on the Public Utilities Commission of Sri Lanka (PUCSL) site, but they painted a rosy picture of a well-functioning grid with improving performance indicators and said nothing about the SCC. Further investigation showed that Cabinet approval had been granted to modernise the SCC with funds obtained from the ADB ($ 17.1 million) in 2012. A tender with a value of Euro 7 million had been awarded to Alstom T&D India in November 2014. 

Even through the new facility was not up and running 15 months after the award of the tender, this proved that my knowledge from 12 years ago was not wrong. The decision to build a new SCC was a formal admission that the old one was inadequate. Why had it taken so long to award the contract? And why was the SCC not operational 15 months later?

The answers came from the CEB’s Additional General Manager (Corporate Strategy) Bandula Tilakasena during a live TV talk show when I brought up the issue. The CEB was aware of the shortcomings of the SCC since 2000, he said. But procurement difficulties had prevented remedial action until 2014. The Director General of the PUCSL said that they had formally asked for remedial action on the SCC in 2010 in the aftermath of a nationwide blackout. 

In a related comment, Tilakasena said that the CEB was aware of the vulnerability caused by having the Biyagama Sub Station as chokepoint in the supply of electricity from the central region into Greater Colombo. They wanted to reduce the vulnerability by extending the grid along the periphery of Colombo, but could not obtain the funds from international development partners such as the ADB. 

Equipment failure in a large and complex system is unavoidable. But resilient design and good procedures should be prevent a simple failure in one sub-station cascading through the system and causing a nationwide blackout. 


The cause

The true cause of the fragility of Sri Lanka’s electricity system can now be discerned. It is the inability of the CEB, a massive enterprise with assets of Rs. 500 billion, to invest in the right things at the right time. The lack of incentives to avoid these kinds of shameful failures is a secondary cause. 

Sri Lanka is a lower middle-income country. We are accustomed to 24/7 electricity unlike residential and commercial customers in the rest of South Asia who have invested in backup power. The economic losses sustained by the country because of unplanned outages are extremely high, ranging from a beauty-care provider who is unable to provide services to her clients to people having to discard medicines because of the lack of refrigeration. Yet, we are still dependent on the ADB’s good will to design and maintain a resilient network.

The CEB is apparently incapable of generating the funds needed to maintain a robust network. The government is apparently unable to give the CEB the needed funds at the right time. Critical procurements drag on for 14 years, but no one is there to take responsibility. Given the slow progress being made on the SCC even after the award of the contract, getting the job done well and on time does not appear to be included in any CEB executive’s key performance indicators.


The cure

The problem is structural. The solution has to be structural too. There is no point in proposing behavioural solutions.

The largest enterprise in all of Sri Lanka (other than the armed forces) should not be going hat in hand to international development agencies for core investments needed to maintain a critical national infrastructure. As a commercial enterprise with massive and assured cash flows, it should be able to raise its own funds.

The structural separation that was decided on by the 1999-2001 Electricity Steering Committee that included all stakeholders should be implemented forthwith. That way, the Transmission Licensee (TransCo) will be able to charge the appropriate rates for the services it provides. It could make the case to the PUCSL that the appropriate rate must include the cost of capital that it needs to raise to build and maintain a robust grid. 

If the Government wishes to support this goal, it can provide funds directly from the consolidated fund or channel concessionary loans to the TransCo. But if these good things do not happen, the Transco will have the freedom to go out and raise its own capital, even at a higher cost. And if it is not allowed to do so, it will have a solid answer the next time the PUCSL asks it to show cause why it failed to maintain supply as required under the conditions of its license.

This is not about privatisation. There is no good reason to privatise a critical infrastructure such as the national grid that has enormous market power in the current single-buyer model. It is about creating a focused and accountable organisation that will not take 14 years from problem identification to award of contract. It is only in such an organisation that executives who are committed to good service can flourish.

Recent columns