Rethinking the Ethiopian privatisation

Introduction

The new Prime Minister of Ethiopia, Abiy Ahmed, has indicated that the privatisation of certain state-owned enterprises (SOEs) is imminent. In August 2018, an advisory board has been established, and the team is expected to get support from legal, finance and accounting experts. However, the Prime Minister has been silent about broader economic liberalisation, including state ownership of land, the 100 or so companies operating as “endowments” and other privately held public interest economic entities. 

Privatisation is an eclectic subject. Economists, political scientists, sociologists, management, finance and accounting scholars have written about it. Pressure groups argue for or against privatisation. Hence, a commentary of this nature has to be pluralistic, and pluralism teaches us that no single theory can explain a complex socio economic and political phenomenon completely. One has to agree that whichever theory one ascribes to there should be enough triangulation of policy, theory or method. New evidence on the privatisation of SOEs has made it clear that the discourse is as much institutional as it is economic (Estrin and Pelletier 2018; Knyazeva, Knyazeva and Stiglitz 2013;Roland 2008). By institutions, we mean laws, organisations, norms, assumptions, values and cognitive matters that are often taken for granted. 

In theory, privatisation has a number of benefits, including the reduction of the burden on the state, improving efficiency, turning loss-making SOEs into viable business entities, improving government revenues from taxes, and serving as a vehicle for attracting new capital and technology into a country. A large body of the literature outlines the determinants of successful privatisation in both developed and developing countries. These gains however are traded off against loss of natural monopolies (e.g. public utilities), and conditioned on a number of institutional factors, including the existence of well-functioning capital markets, an effective regulatory environment that opens markets for local competition as well as adherence to sound corporate governance standards and control of corruption.  

Privatisation in transition and developing economies ended up transferring public assets onto the balance sheets of foreign companies and into the coffers of politically connected individuals (businesses) under shoddy deals. They failed to liberalise the sectors, hence essentially replaced state monopoly with private monopoly. Had the transferred assets valued correctly, transaction costs reasonable (market efficiency) and the cash raised from the sale used for the wellbeing of citizens who gave up benefits from public ownership of the SOEs, then privatisation would have little or no distributional effects. The tax base would have expanded had the buyers, sellers and financiers been domestic residents. Academic studies show that the gains in the investment efficiency frontiers of privatised SOEs had been difficult to measure, and they are often traded off against losses due to under-pricing, transfer price malpractice, corruption and a de facto retention of monopoly rights in the post privatisation period.  Where the gains are measurable, they were limited to specific sectors, namely in the banking and telecommunication industries. The reforms have had adverse effects in that consumers, infant industries and workers were unprotected. In short, the institutional aspects of privatisation are important to make the promises realisable. Despite the mixed evidence, both the World Bankand Privatisation Barometer Report 2015-16indicate that governments all over the world continue to transfer public assets into private hands in record fashion and raise substantial amount of money from the sale.

Privatisation and liberalisation 

Privatisation of SOEs is not an isolated incidence. It ought to be one of the cornerstones of liberalisation, and liberalisation in turn entails a package of sequenced reforms aimed at improving an economy and reducing uncertainty. Liberalisation succeeds or fails depending on the degree of investor protection, quality of laws and its enforcement and the pre-existence of deep and functioning domestic capital markets. For a developing country, liberalisation may include taking a series of steps such as modernising a country’s commercial and security laws aimed at improving the depth and breadth of its financial markets, protecting investors, and establishing a settlement system that ensures liquidity and credible reduction of information asymmetry. 

A credit bureau that stamps out bad borrowers and “bank looters”, the existence of independent and credible analysts and audit firms that distinguish bad business and securities from good ones are important as well. Furthermore, a regulation that disallows political parties from owning, investing and controlling companies should be erected as they create not only distortions in product and labour markets, but also increase corruption and interference in elections.

It is also important to note that liberalisation has its own winners and losers as the market re-assesses the risks and returns of companies and sectors. At a more general level, the unique features of observed liberalisation programmes include ownership rights rearrangements, deregulation, trade liberalisation, currency adjustments, removal of foreign exchange restrictions, rationalisation of government bureaucracy and finances, creation of an effective social safety net, and lower spending in social investments (education and health). 

In short, privatisation is an important component of liberalisation, and liberalisation is contextual as it comes in different forms, packages and sequences. A cabinet that privatises SOEs without some degree of liberalisation is like a board of a company that decides to sell investments in plant and machinery only to relieve itself from an impending liquidity crisis. In other words, privatisation with the aim of alleviating short-term cash/foreign exchange/ constraints is not going to be a good proposition.   

Privatisation in Ethiopia 

However, at slow start by African standards, the Ethiopian People’s Revolutionary Democratic Front (EPRDF) government has had at least two waves of privatisation. The motivations for each wave however remain unclear. For instance, Getachew Regassa (2003)[[i]] indicates that the government privatised a number of hotels, farms and small factories primarily through direct sale in the 1990s. The second wave was on mines, breweries and long-term lease of urban and rural pieces of land. Despite these waves of transfer of public assets into private hands, Ethiopia’s big business sector is dominated by SOEs, political party controlled and managed enterprises, beneficiaries of the entrenched clientele-ism and a few genuine companies. The government of Ethiopia has had no firm policy on SOEs and party controlled and managed businesses.  In some years, the ruling party was observed selling SOEs while in other years the EPRDF government resisted privatisationand deployed big public resources on SOEs, often financed by public debt, under an ideological mantra of “revolutionary democracy and developmental state”. 

In November 2014, when Ethiopia went to borrow from open commercial debt market, we used a commentary entitled “Is Ethiopia’s sovereign debt sustainable?” in order to document the risks associated with over borrowing from the open market, and advised that the country rather diversifies its sources of external finance and continue to look for concessionary and bilateral loans. In that commentary, consistent with theory, we argued that debt is acceptable if the rate of return from a new project is greater than the financing costs.  

We explored additional critical issues such as the governance and management of the national debt; the behaviour of credit rating agencies; the then financial flows into sub-Sahara Africa; the debt sustainability measurement method; stagnation of exports (debt to export ratios); the ways in which the government guaranteed SOEs’ debts and related matters.  We fear that borrowed funds that were invested in hydro, sugar, railways, industrial parks, etc. have produced negative returns, cost overruns, delays, shoddy works and cash starved mega projects. It appears that the debt burden is now worse than it was in 2014. 

The 2015-2018 period was a period of unrest that exacerbated low intensity conflicts, communal violence and revanchism, forcing the regime to spend even more on security and declare two state of emergencies over the entire country. It was also a period of high uncertainty, causing increased capital flight, outmigration and unprecedented level of internal displacement. The sustained unrest increased the gap in the fiscus: it decreased revenues from exports and remittances, thereby precipitating schism within the ruling ethnic coalition established through political and economic cronyism, patronage and clientele-ism. Like the saying goes, every crisis starts at the treasury, what we are hearing from the reformers within the party is that the country is under severe financial distress. 

They appear to be obliged to fill the short fall though new borrowing (United Arab Emirates), debt restructuring from the main lender (China), sale of public assets, and appealing to the conscience of the Ethiopian diaspora. As of late we are hearing allegations that megaprojects were used to plunder public resources, contracts and land were awarded without transparent tender processes, often through personalised intermediaries or guanxiand shell companies in lender countries, suggesting the need for special reforms to introduce accountability and transparency.  In other words, the mega projects have not raised the expected returns to pay off the principal and the interest. All these mean that the post-conflict transition period is not a walk in the park as the new Prime Minister has a lot on his plate.  

While the Prime Minister has correctly identified the efficiency and accountability problems in SOEs, we suspect that the primary driver for the privatisation project has to do with debt and general foreign exchange shortage, and agency conflicts appear to be either unknown or secondary. The absence of a carefully sequenced economic liberalisation and stabilisation programme suggests that the Prime Minister is overwhelmed with crisis management, handling the factions within the party that threaten the unity and territorial integrity of the country, and ensuring internal and regional stability in the rapidly changing Greater Horn of the African region.  

All of these actions are undoubtedly important. The absence of meaningful transformation of the cabinet that generates timely reform programmes for example by generating laws that regulate/re-organise party owned companies raises legitimate doubt about the depth, rigour and sequence of the liberalisation agenda. In other words, the political liberalisation observed during the last few months needs to be matched by a carefully designed economic liberalisation and stabilisation programme. 

One may ask a simple question as to why the government is planning to embark on selling public assets when there are political party-owned companies that are not only inefficient and ruling party’s zombies but also distort both the product and labour markets, and adversely affect the promised democratisation process. Moreover, it appears that the government did not take stock of the country’s past privatisation experience before undertaking a new wave of privatisation. In other words, the absence of separation of powers and the lack of decoupling the state, the party and the economy have created a unique form of networked capitalism that captures and weakens the effectiveness of the regulatory and supervision machineries of the country, thereby legitimising the continuation of state capture. 

Thus, privatisation in a networked or captured economy of course transfers public assets in disguised and opaque ways to those who are already connected. Privatisation without institutional reforms may only relieve the government from temporary illiquidity but does neither affect the efficiency nor make the market contestable nor prevents throwing away public assets at questionable prices.

We suspect that the government is contemplating privatising its “crown jewels” like Ethiopian Airlines and Ethiopian telecom, both of which enjoy monopoly powers. Unlike many African airlines, the national airline has been successful, and as of late, it is acquiring smaller African airlines, which makes it an acquisition target. This is despite empirical studies on mergers suggesting that the long-term benefits of business combinations for acquiring companies has not been great as the anticipated synergetic gains stemming from the merger are difficult to realise. 

We also note that, thanks to the leasing industry (collaboration between manufacturers and finance houses), many smaller airlines work with leased equipment, and cash flows are more from landing rights, and the Ethiopian Airlines is not an exception. The removal of state guarantee for its foreign debt and making government business competitive could lead the SOE into the same unfortunate situation as the South African Airways finds itself. 

Furthermore, the new lease accounting (FASB 842 and IFRS 16) may unravel the effects of “off balance sheet financing”. While the strategy of the airline in promoting intra-African investment and communication appears sensible, it is important to examine whether the cost of such endeavour could be absorbed by the rest of the Ethiopian economy, without creating major economic dislocation. Yet Ethiopian Airlines has proved itself to have operational, knowledge and large diaspora market in a globally competitive air transport market. It is therefore critically important to be selective in allowing what kind of investors (financial capital) Ethiopian Airlines invites. The method and modality of privatisation of Ethiopian Airlines is as important as the privatisation itself.    

In our opinion, the real test for the national airline is not just partial privatisation through private placement that recycles oil dollars or looking for loan or equity partners from the east. Its strength lies in capitalising its efficiency by reorganising itself for an eventual listing in a credible global capital market. In contrast to the national carrier, the strategy for the Ethio-Telecom might be different. For instance, it is worthwhile to look into how the South African telecom market was liberalised and reorganised. The monopoly was broken down allowing Vodacom, MTN and Cell C to serve the market. 

This form of breaking up monopolies is somewhat similar to what Ethiopia did for its banking and insurance sectors. What is missing is an organised capital market and updated commercial and securities laws. Filling this gap could pave the way for financial deepening and the privatisation and reorganisation of other SOEs and political party-owned enterprises. It is in this light and with the above caveats that we agree with the Prime Minister that SOEs that do not serve as “commanding heights” or enjoy “natural monopolies” can be privatised, systematically. 

With regard to the methods of privatisation, there are several ways of disposing state assets. Direct sale or tender, private placement, equity swaps for exiting/new loans, and Initial Public Offerings (IPOs) have been used to privatise all or parts of SOEs. In some countries, blocks of shares were sold at discounted prices to trade unions, war veterans, civil servants and other designated groups that deserve restorative justice. Direct sale by far is more likely to lend itself to mispricing, bidder collusion, abuse of investment incentives, corruption, and makes bargain hunters the winners of privatisation, allowing them to earn super profits without requiring them to bring bricks and mortars to the economy or spending in research and development. In other words, governments must not only see the foreign currency inflow from the sale.  

Concluding remarks

Notwithstanding the series of failed elections Ethiopia experienced over the last 27 years, we believe that Prime Minister Abiy Ahmed has considerable public and international support, which could be used to put the country on a stable transitory path.  In the sphere of economic policy, the government needs to undertake a comprehensive assessment of economic policies and identify specific areas of reforms. While the policy statement on the privatisation is prudent as an indicator of an overall policy orientation of the reform package, several issues need to be addressed before actual privatisation. These include, among other things, the need to understand that the political costs of privatisation in a period of post conflict election could be relatively high and there is need for a series of regulatory and institutional interventions.  

To conclude, privatisation, whether it is a complete divestiture, partial sale, private placement, debt-equity swap, or IPO requires proper and effective regulatory framework, creation of new institutions, the selection of a reputable international investment bank, an innovative valuation and market timing. Revenue raised from privatisation, as the one reported in Privatisation Barometer must not be the only incentive for the upcoming Ethiopian privatisation. All of the schemes have advantages and pitfalls. Overall, privatisation without changing the institutional framework would not bring about improvement on the performance of SOEs or attract new investment. The best that one can get is new cash from the disposition of public assets. In the sphere of economic policies, it is time to consider the sum total of the motivations for privatisation and determine whether this should be done now or after the election.  

* Minga Negash is Professor of Accounting at Metropolitan State University of Denver and the University of the Witwatersrand; Seid Hassan is Professor of Economics at Murray State University; Abu Girma is Associate Professor of Economics at University of Tsukuba; and Ezana Kebede is an automation solutions consultant for banks in Ethiopia.

* The opinions expressed in the article are those of the authors and not of the institutions they are affiliated with.