Something is fundamentally rotten in SA's nine state-owned enterprises (SOEs). Most have had chronic governance problems, leading to weak leadership, operational confusion, financial uncertainty and a failure to deliver. They have to deal with an investment backlog, political interference and a globalised economy that severely punishes strategic lethargy.
Government has finally admitted there's a problem (though not that it is the problem). President Jacob Zuma said this week that a serious review of SOEs was needed. "Certainly we cannot continue with them like that. It's a problem with the leadership and we are looking at that."
Ewald Muller
The danger is that a review will look in the wrong places. It would be all too easy explaining the problem by reducing it to the level of personal duels - as in the recent confrontation between Eskom chairman Bobby Godsell and CEO Jacob Maroga (both now former). But that wouldn't explain why the four most important SOEs have all been plagued by executive instability in the past year.Take Transnet. Four key positions are filled in an acting capacity: chairman, group CEO, chief financial officer and head of Transnet Freight Rail, the parastatal's most important division. Eskom has the same man, Mpho Makwana, acting as chairman and CEO, and it appointed a financial director last month after the post was left vacant for all of last year.
At SAA, Chris Smythe has been acting CEO for a year, and for much of last year the SABC had no CEO and its chief financial officer was suspended for three months. Also in the past year there has been almost complete turnover in the boards of SAA and the SABC (twice). Even where the boards have been relatively stable, directors have been mostly powerless to perform their main functions: the appointment of the CEO and other top executives, and then holding them to account.
This paralysis would simply not be tolerated in the private sector, where succession planning is generally thorough and acting appointments are rare and brief. That's because they lead to strategic uncertainty, operational inertia and a loss of confidence among investors and stakeholders.
But private-sector companies are run by boards which answer to shareholders, with no doubt about who is accountable to whom. There is no such structural clarity in the state sector.
The boards of SOEs are appointed by government, through the public enterprises minister. But government also in effect reserves the right to appoint the CEO, which bypasses the board and blurs accountability. At worst, the board ends up holding great responsibility but with its powers diluted to the point of impotence as it tries to please the "shareholder".
This was demonstrated in the Eskom board's clear dereliction of duty when government failed to take its advice on the need to build new power stations. Ian McRae, the former Eskom CEO who founded and headed the national electricity regulator from 1995 to 1997, had approved a power investment plan. He noted last year that the new generating capacity would have come on stream in 2006: "Eskom would have been spot-on had government not interfered." McRae has also attacked the Eskom board for not only succumbing to government, but failing to warn the country of the risks in delaying the investment. This would have alerted customers and possibly forced government to treat the matter with more urgency.
The inherent governance flaws are made worse by government seeing in SOEs opportunities for patronage (placing loyal ANC members in lucrative positions) and high-profile black empowerment. Both imperatives, which would help explain the long delays in making permanent appointments, seem to have taken on the status of virtually unbreakable rules. It does not follow that those appointed are necessarily unsuited or unqualified, but the risk is there. Ideological confusion over the role of SOEs has added to the malaise.
In the early 1990s, the concept of partial or full privatisation was fashionable. As late as 1996, the partial sale of Eskom power stations was being considered. Sivi Gounden, then public enterprises DG, said government wanted "to encourage maximum private-sector participation while also maintaining a level of control in Eskom". Then the mood shifted back to heavy emphasis on the SOE as an instrument of social development that could not be left to the market.
In the past year the pendulum has begun to swing back, with Eskom considering partnerships with private players in generation and Transnet willing to concession low-density branch lines. But the uncertainty meant investment and maintenance were often put on hold while managers waited for clarity.
Less obvious, but no less destructively, the effectiveness of the SOEs has been steadily undermined by the financial models they are obliged to use. There was a realisation in the 1970s and early 1980s that the SOEs could not continue to be run like government departments. Though Eskom and Transnet had a solid record of delivery and pockets of management excellence, they were also inefficient and overstaffed (the railways had been used as protected employment for those of the white electorate on the economic margin ). Their funding models were inadequate, and individual advancement depended largely on seniority.
Corporatisation resulted in improvements in oversight, reporting, financial controls and executive compensation. In the 1980s, Eskom became a shining example of what was possible.
But the unqualified adoption of private-sector financial principles and systems proved to be something of a virus rather than a cure. The first mistake was the assumption that SOEs should be driven by a profit motive. This is nonsensical. Their only shareholder, government, has a source of guaranteed revenue (tax) that is not available to the private sector. Philosophically, the whole point of an SOE is that it provides a service at the lowest possible cost, for the economic benefit of the country. It should not be expected to make a profit.
What an SOE should do is run its operations in a business like manner so that, ideally, it does not have to be subsidised by the taxpayer. If it generates an excess of income over expenditure, that should be regarded as a surplus to be reinvested. And if the surplus is excessive, then it can be used to cut tariffs and therefore the cost of doing business. This is a powerful rationale for state ownership but it is undermined by the emphasis on profit.
Some private-sector financial measures are obviously useful for SOEs. But the danger is that executives are rewarded for getting priorities wrong.
At Eskom, for instance, the power crisis in January 2008 was triggered by coal shortages: managers had been rewarded for running down inventory (a financial measure) rather than maintaining a stockpile (operational). In August 2008, in its review of the January power crisis, Nersa noted that "there was a reluctance to obtain supplementary coal due to its high cost and its impact on Eskom's financial position".
The lesson is clear: wrong incentives encourage wrong behaviour and divert funds to unproductive use.
Transnet, under Maria Ramos from 2004, handled its finances and capital expansion plans better than Eskom did. But in the income statement in the 2009 Transnet annual report, there is a reference to R7,76bn of "profit attributable to equity holder". This does not mean Transnet actually paid that amount to government, but it represents 23% of revenue. Those customers who are moving from rail to road can be forgiven for wondering if Transnet is doing its best to keep down tariffs.
The other principle that is dubious to the point of absurdity is that SOEs are expected to pay tax. In 2009 Transnet was subject to an effective 17,75% rate, which translated into a payment of R1,64bn (5% of revenue). This simply means that government is taxing itself. Put another way, it is an extra tax on the country. This money would be better employed in Transnet's hungry capital expenditure budget, or on cutting costs.
"Tax for SOEs is nonsensical," says SA Institute of Chartered Accountants senior executive Ewald Muller, "and dividends should occur only in rare circumstances, probably only when there are capital profits that are real."
Muller does point out that the IFRS accounting reporting framework, including profitability, is needed when entities like Eskom and Transnet go to global credit markets for major funding - "but not as a management incentive".
However, Transnet's own choice of key performance indicators (KPIs) is not governed by IFRS, and reflects a financial rather than an operational mind-set. Seven of its nine KPIs are financial (such as Ebitda margin, cash interest cover and cash flow return on investment), with only two - rail maintenance and safety targets - relating to what actually happens in the business.
Muller believes that "different metrics should be used by SOEs - for example, return on assets employed, cost:revenue and cost:staff numbers".
An irony is that when Eskom was converted by legislation into a public company in 2001, the general assumption was that this was the first step towards privatisation. At the very least, said a public enterprises department spokesman at the time, conversion "paves the way for its restructuring". Government said it was committed to the "managed liberalisation" of energy, retaining Eskom as a "dominant" supplier. Nearly 10 years on, Eskom is still a state monopoly, a vertically integrated dinosaur. But in that time it has paid an estimated R12bn in company tax and R1,11bn in dividends to the state.
Another effect of the conversion was the abolition of the successful two-tier structure of Electricity Council and Eskom management board, which were replaced by a board of directors appointed by government. This apparent streamlining made it easier for government to interfere and "get into the engine-room", as one former Eskom executive puts it.
The diversion of Eskom money to government did not provoke much opposition at the time. The arrangement seemed merely a step on the route to privatisation, and it was widely taken for granted that Eskom would continue to provide cheap electricity and generate a surplus. It was also assumed that the infrastructure would remain adequate (Eskom knew better, but did nothing to contradict this cosy impression).
Despite all the fancy reporting standards, most of the SOEs have had to be routinely funded by government, bailed out, or provided with guarantees to enable them to borrow.
Eskom is dealing with at least a 10 -year investment backlog, and Transnet with 30 years of under investment and poor maintenance. The SABC has been badly mismanaged and its business model has always been self-contradictory. SAA has suffered from strategic mistakes, extravagant management and hostile market conditions.
But there can be no doubt that the governance and financial models of SOEs have made things worse - and, unless modified, will continue to block good business practices. In the case of Transnet and Eskom, this could amount ultimately to sabotage of the economy.
What has to be done? The first step would be to acknowledge that just about the only thing that the SOEs have in common is that they fall under the public enterprises minister. They have different histories, different problems and different roles to play in the economy. There is no one-size-fits-all solution.
Even government now seems to accept that Eskom's monopoly needs to be broken. It is both supplier and buyer of electricity, and there is no business case for generation, transmission and distribution being controlled by the same entity. This model exists virtually nowhere else in the world.
It may well make sense for Transnet, on the other hand, to remain an integrated rail and port transport chain. But what is the rationale for it to continue operating the Durban-Gauteng oil pipeline?
Once part of Transnet, SAA now seems to be a prime candidate for total privatisation (as is the smaller SA Express airline). However, state ownership of SAA was defended by former public enterprises minister Alec Erwin as "strategic". This was never explained.
It might be argued that a country like New Zealand needs a national airline because it has few people and is far from everywhere: big global carriers might find it uneconomic to provide a service. Was this what Erwin had in mind? What else could be meant by "strategic"?
There have always been additional political sensitivities influencing the governance of the SABC (which reports to the department of communications rather than public enterprises, but it is also undoubtedly an SOE). Politics has at times come close to short-circuiting the broadcaster's ability to perform its basic functions - hardly surprising, given that cabinet has the power to appoint the CEO, chief financial officer and chief operations officer.
More deep-rooted is a structural tension because of the insistence (going back to the 1970s) that public-service broadcasting should be subsidised by commercial activities (which, absurdly, was made more difficult when some of the SABC's cash-generating radio stations were sold off in the 1990s).
Structurally, there are only two realistic ways forward for the SABC: continue the cross-subsidisation, or privatise the commercial stations and pay for public broadcasting openly out of taxes. Meanwhile, because the SABC has experienced the worst management of all the SOEs, much work needs to be done to restore routine efficiencies and staff morale. That focus is necessary - but it will only defer solutions to the bigger political and funding problems.
As for the smaller public enterprises, there is even more confusion. Denel, the rump of the once-mighty Armscor, has been losing money, not least because it has not been given much support from the SA National Defence Force's procurement process - hardly a gesture of confidence.
If an arms manufacturer is not an extension of a country's military (as Armscor once was), why should the state be involved in that industry at all? By contrast with Armscor, Denel is too small to be strategic in the sense of being a supplier of last resort for SA, and it lacks the critical mass to be globally competitive except in niche areas. It should be sold.
So should Alexkor (diamonds) and Broadband Infraco (telecommunications), each in markets well served by private companies which understand risk and reward. Safcol (forests) is earmarked for disposal.
By contrast, the ninth SOE, Pebble Bed Modular Reactor (PBMR), is an innovative project to increase the generation of electricity through nuclear power, and it has great strategic value for SA. Yet it appears to have been hobbled financially by its corporate association with Eskom.
In principle, PBMR is the best kind of public-private partnership, with investors ranging from the SA government and Eskom to the Industrial Development Corp and US nuclear giant Westinghouse. Nearly 40% of PBMR employees have honours, master's or doctoral degrees, skills that SA badly needs to retain. The PBMR deserves to be supported but it does not need to be classified as an SOE.
WHAT IT MEANS
 There is no one-size-fits-all SOE solution
Public enterprises ministry is redundant
|
Full privatisation of SOEs is not new, of course: the previous government took the plunge with both Iscor (now ArcelorMittal SA) and Sasol. And partial privatisation was tried with Telkom, which is listed on the JSE but with the state as a major shareholder.
If the Iscor/Sasol route is too much to hope for, Telkom may seem to show the way for some of the remaining SOEs. But Telkom is neither fish nor fowl. Government does not need the rewards that derive from holdings in listed companies, nor should it be exposed to the risks. And if it feels it needs to influence a sector for strategic reasons, it can do so quite adequately with regulation.
While government mulls its options, it needs to understand that nothing will be fixed until the flaws in SOE governance and the financial models are acknowledged.
But first prize would be to hear public enterprises minister Barbara Hogan announce that her aim is to legislate herself out of existence.
- See also Budget 2010 supplement
Ranking the Analysts 2009
