The basis of historical evidence

Part II of the three part series against privatisation policy in Pakistan. The first part can be accessed here. Part II appeared in The News on Sunday on December 01, 2014 (http://tns.thenews.com.pk/the-basis-of-historical-evidence/)

The Pakistan Muslim League-Nawaz government has listed some 31 state-owned enterprises (SOEs) for privatisation in the near future. There is no information available at the moment about the modalities of the process but, most likely, it will be along the lines of the procedure followed at the time of privatising Pakistan Telecommunication Company Limited (PTCL) when the government technically only ‘partially’ privatised the SOE but gave up its control over the management of the company. This is how it works.

Government offers 26 per cent shares with management control in these companies to private investors and retains a maximum of 74 per cent shares with it. The management control is given by writing down in the contract that each 1 per cent of the privately owned share carries 4 votes (as opposed to 1 vote for each percentage point of the remaining 74 per cent) in the board of directors (BoD) of the privatised company. This way, the minority shareholder private company (or consortium) commands 104 votes in the BoD and hence also manages the daily and other affairs of the company.

Governments usually face little difficulty in privatising profit-making enterprises. Selling loss-making enterprises poses a much difficult challenge. This is why each privatisation phase is preceded by a restructuring phase. This is the time period during which the SOEs are primed or dressed up before they are sold off.

This is done to ensure the future hiring and firing and investment decisions are in line with the aims and objectives of the new private owner-cum-manager.

Let us now focus our attention on what the evidence actually suggests. The frequently quoted success stories of privatisation in Pakistan are the banking sector and the telecom sector. In electronic and print media, their profitability is presented as an irrefutable evidence for their success. We earlier argued that privatisation argument is made on efficiency grounds.

Let us see what various studies done on the performance of the banking sector in Pakistan say about this indicator. The conclusions give one many reasons to be more humble in claiming undeniable virtues of the privatisation policies. Akhtar (2002) studies the efficiency of privately-owned, public-owned, and foreign banks in Pakistan for the year 1998. He concludes that in terms of efficiency, privately-owned banks do better than the other types and therefore give support to the privatisation argument. A closer look at his results suggests that such a strong conclusion is unwarranted.

Akhtar develops a mean score for efficiency for all three types of ownership structures. The average score for overall efficiency for privately owned banks is 0.80, for publicly owned banks is 0.77 and for foreign banks is 0.75. A difference of 0.03 between privately- and public-owned banks forms the basis of Akhtar’s conclusion.

This is unwarranted because it requires that one must clearly show this difference is entirely due to the difference in ownership structure and for no other reason. What also needs to be shown is that the difference is significant to warrant the strong conclusion drawn from the analysis.

Akhtar’s analysis falls short on both these counts.

Qayyum and Khan (2007) measure the efficiency in the banking sector from 1998-2005. They have not directly focused on changes in efficiency due to ownership structures but conclude that efficiency is falling for the domestically-owned banks since 2000.

Akmal and Saleem (2008), in their study on the banking sector from 1997-2005, found the opposite result. They argue that efficiency of the banking sector has generally improved since 2000. They also find that, during this period, the efficiency of the state-owned banks has improved the most. Again, the improvements are only in the relative sense and not against some international benchmark.

Burki and Niazi (2006) is the only study I have read that finds non-state owned banks to be more efficient than the state-owned banks. However, Ahmad (2008) used a methodology similar to Burki and Niazi’s to conclude that the state should not promote privatisation of state owned banks!

The point of the discussion is that it has been quite difficult to show that privatisation policy (coupled with liberalisation) was the key ingredient that improved the performance of the banking sector. The studies that have attempted to show this have found results that either contradict each other or tend to oppose the results advanced by economic theory.

The privatisation of Pakistan Telecommunication Company Limited (PTCL) is also roundly hailed as a success story. I do not feel the need to go into a detailed analysis of what various studies have concluded regarding PTCL’s performance in the post-privatisation phase. I believe it is sufficient to make two points here. First, I urge the readers to read Kamal A. Munir’s excellent piece on PTCL’s post-privatisation performance that appeared in The Express Tribune in 2012. Second, the new ‘owners’ of PTCL still owe the government 800-900 million US dollars since 2004/05.

The previous government made considerable efforts to recover this amount but to no avail. Let us wait and see how the current government led by Nawaz Sharif fares on the issue.

There is another important point to note. Governments usually face little difficulty in privatising profit-making enterprises. Selling loss-making enterprises poses a much difficult challenge. This is why each privatisation phase is preceded by a restructuring and reorganising phase. This is essentially the time period during which the SOEs to be privatised are primed or dressed up before they are sold off. This particularly requires firing workers or weakening the unions in order to “rightsize” the enterprise.

Prospective private investors/buyers usually require the government to undertake these actions to avoid political costs of rightsizing an enterprise after privatisation. But this phase does not just involve such sensitive political decisions but also at times requires substantial amounts of investments to turn the enterprise around. It is obvious that all government outlays in this turning around phase come from the public.

This raises an important question – why must public resources be utilised for eventual private gains?

In this round of privatisation, these issues will become particularly important. For example, the most cited figure to highlight mismanagement in Pakistan International Airline (PIA) is that its employees-to-aircraft ratio is over 450 employees per aircraft when the international average is between 150-180employees per craft. Due to this mismanagement and other reasons, the only way to sustain PIA is to privatise it. PIA currently employs around 19,000 employees. If it has to be privatised it must be primed up before the final transaction.

The politically sensitive decision of rightsizing the organisation before privatisation will have to be taken by the government. This means that a maximum of around 11,000 works needs to be fired. This is a figure that you and I don’t hear about from our economic experts.

Similarly, PIA is taking decisions to buy new planes or upgrade existing ones. All of these will be paid for by the public money and, after privatisation, passed on to the private owners.

Similarly, the other SOEs that all our experts seem interested in privatising are electricity distribution companies (DISCOs). This is not going to be an easy task, particularly because of the way WAPDA’s grid is organised. This may require substantial investments financed by public money.

The problem is that WAPDA’s grid has evolved in such a way that it is difficult to ensure that power plant A provides electricity only to City A. So if a distribution company is privatised, the owner would demand some sort of guarantee regarding the supply of electricity. Some basic amount will have to be guaranteed to it, otherwise no buyer would be willing to touch a DISCO. This became a crucial issue in Karachi Electricity Supply Corporation’s (KESC) post-privatisation performance.

KESC found it difficult to supply electricity at an affordable price using its production plants. Ultimately, a special agreement had to be signed between KESC and WAPDA that promised the former around 600 MW of electricity on a daily basis. But WAPDA’s electricity is subsidised to ensure its accessibility to the public.

WAPDA sells this subsidised electricity to KESC – a private distributor. KESC in turn sells this electricity at the rates that are determined on the basis of its own cost of generation. In such an arrangement, what is essentially happening is that the public at large is subsidising the private owner of the company. This is not entirely a new thing. Pakistan has a long history of government creating rents through its various policies that ultimately became channels of wealth transfer from the public at large to a few private investors.

Hence, whenever any DISCO is privatised in the future, the government will have to ensure some minimum supply of electricity. This can happen only in the long run where the power production is increased gradually in the country. Again, the state’s role in it will be crucial. In the absence of any supply guarantees, the government will have to offer contracts to the buyers of DISCOs that promise a certain rate of return on their investment.

The point here is that privatisation of DISCOs cannot work unless the government undertakes major restructuring of the power sector and offer lucrative contracts to the private owners. This requires considerable levels of investments, which will most likely come from the national exchequer or publicly-guaranteed foreign debts. The question is that if the government must undertake such expenditures, then why should we hand it over to a private investor? The answer to this question has important ramification that I shall discuss in the next and the final part.

(to be concluded)