The very purpose of most state-owned enterprises (SOEs) is to sidestep competition. The ideal outcome for enhanced trade and investment and global economic performance is thus a ban on SOEs from as many sectors as possible, and this should be the long-term American goal.
A ban on SOEs, unfortunately, cannot be accomplished with a single free trade agreement, and may not be entirely achieved with a series of trade agreements. Countries must come to see the benefits of limiting SOEs for themselves, which will take a good deal of time. An alternate concept for the short term is “competitive neutrality”—fair, undistorted competition between SOEs and private firms. It is an appealing goal, but some variants of competitive neutrality are seriously inadequate, relying on vague prescriptions that bind only countries eager to implement them, not those looking to exploit loopholes. The result is antithetical to good trade policy.
The proper American objective concerning SOEs for the Trans-Pacific Partnership (TPP) is to retain the objective of competitive neutrality, but overhaul the inadequate versions that have been offered publicly to date. The American negotiating position should be clear and coherent: Only a genuinely useful variant of competitive neutrality is acceptable. Otherwise, given the importance of SOEs, the TPP will be undermined in terms of substance and thereby in the eyes of Congress.
The Growing SOE Challenge
State firms have been growing in importance for the past decade. The 30 largest companies in the world in 2012 included 12 SOEs from eight countries. The biggest companies in TPP members Malaysia, Mexico, Singapore, and Vietnam are SOEs, as are the second-largest and third-largest companies in TPP member Japan. In Brunei, another member, the largest company is half-owned by the state. Looking ahead to a potential trade agreement between the U.S. and the European Union, the EU is still home to various state giants, such as Italy’s oil and gas corporation Eni. Perhaps most important in the long term, Chinese SOEs have become a model for many developing countries.
In addition to the distortions it imposes on individual economies, increasingly prominent state ownership is antithetical to liberalizing trade accords. State ownership, deliberately and otherwise, circumvents or neutralizes international economic obligations. Numerous members of both parties in the U.S. Congress have made clear the importance they attach to limiting SOEs in U.S. trade agreements. A TPP with a shallow form of competitive neutrality would be at risk.
Aggressive treatment of SOEs discomforts interest groups in many countries, not just those with obvious state sectors. Singapore has two sovereign wealth funds, a special form of financial SOE, whose influence over the rest of the economy is excessive. In the U.S., a variety of interests protect what should be seen as SOEs, for instance, firms recently saved from bankruptcy by state action. Hence, the search for a way to address SOEs while avoiding domestic pain—the most popular answer to which has been competitive neutrality. However, current forms of neutrality have plainly not inhibited SOEs.
Competitive Neutrality to Date
Competitive neutrality was first articulated in Australia in the 1990s. It is a voluntary and laudable decision by the Australian government to limit the anti-competitive influence of SOEs within Australia. It is intended, properly, to “eliminate resource allocation distortions arising out of the public ownership.”
Remedies are only to be applied when the benefits of doing so are believed to outweigh the costs. But state entities presumably seek ownership precisely because they saw some net benefit, whether real or imagined. This creates an immediate presumption that no action should be taken.
Seventeen years after its adoption it was still a struggle to apply the notion of competitive neutrality to as prominent an entity as the National Broadband Network. Even limited judgments were rejected by the government because “higher prices or lesser services would be necessary in rural and regional Australia and the alternative of funding of rural subsidies from the budget would create permanent uncertainty.” There is a benefit for the network, so the requirement of competitive neutrality should be waived. And this was at the national level, where inconsistencies can be most easily seen.
Application of neutrality is at greater risk at lower jurisdictions due to lower visibility and a far larger number of cases. Australian policy “gives each jurisdiction responsibility for determining how, when and to which businesses, competitive neutrality should be applied. For example, each jurisdiction is responsible for determining those businesses where the benefits of applying competitive neutrality outweigh the costs.”
Perhaps most telling, Australian competitive neutrality emerged within a generally market-driven environment. As such, it was aimed principally at financial subsidies, as these were most salient. Australian neutrality is not designed to address anti-competitive regulatory preferences for SOEs because these were not a major issue. However, they are a major issue elsewhere.
Australian competitive neutrality was thus not designed for internationalization, either through internal use in other countries or in transnational economic agreements. If a national government that voluntarily chose neutrality dismisses it when inconvenient, as with broadband services, local authorities in ambivalent or opposed countries will not be reliable practitioners.
Competitive neutrality is nonetheless being modified for internationalization by the Organization for Economic Cooperation and Development (OECD). The OECD treads heavily on the ground broken by Australia. It cites separating commercial and non-commercial activities to the extent that benefits outweigh costs, again providing a ready excuse not to act.
The OECD rightly emphasizes transparency in evaluating commercial and non-commercial activities, praising Norway’s performance on this score:
All SOEs are officially designated as being either policy oriented (i.e., having primary objectives other than profitability) or commercial. The latter category is sub-divided into three categories, namely (i) fully commercial; (ii) commercial but with an obligation to maintain headquarters in Norway; and (iii) commercial but required to pursue certain additional objectives.
This clarity reveals the obstacles to achieving neutrality. Either the primary objective is not commercial, or a set of secondary objectives is not commercial, or the SOE is commercial save for an anti-foreign bias. The case where the SOE is “fully commercial” is extremely unlikely, requiring that an SOE can fail and not be replaced. Moreover, the OECD admits, “It would appear that, perhaps reflecting the high degree of transparency around these procedures, political considerations sometimes play a direct role in the categorisation of SOEs.” Transparency reveals that governments highly value non-commercial SOEs.
The OECD recommends that SOEs receive “adequate compensation” for the public policies they must implement. One country’s adequate compensation is another country’s heavy subsidization. There are instances where countries defend compensation for SOEs as a form of public investment. Regardless of the terminology and justification, such transfers violate any reasonable notion of neutrality.
The OECD stresses separation of market regulation from SOE management. It insists creditors have equally powerful claims against SOEs as private firms. But when the independence of government arms, including the judiciary, is compromised by single-party rule or other forms of political centralization, neither the formal separation of regulator from SOE nor creditor rights can have real meaning. This flaw is particularly important when the regulator is mandated with disciplining SOEs that make purchases and sales on a non-commercial basis, disregarding market supply and demand.
There are many examples of OECD prescriptions being far too vague or simply inapplicable. Country variations mean internationalization is a challenge even when all countries do have the desire and the capacity to act, as in the OECD. For countries with extensive state sectors, such as Malaysia, Vietnam, and undoubtedly others, OECD neutrality guidelines are utterly inadequate.
It would require a sweeping overhaul of these versions of competitive neutrality for it to be a sound component of the TPP and similar agreements. Beyond the problems with internationalizing neutrality, there are more fundamental shortcomings.
On a strict interpretation, competitive neutrality cannot truly exist. Firms that do not operate on market principles, say, because they retain surplus workers during a downturn or undercharge for oil, cannot survive in competition with firms that do. They must be subsidized. The most basic dimension of corporate competition is entry and exit into the market. SOEs are created precisely because governments want corporate agents that do not operate according to market principles. They exist, in this sense, for the opposite reasons private-sector firms exist.
The exit side is equally stark. For competition to be fair, all parties must be equally capable of bankruptcy. State firms almost never go bankrupt. They are privatized (and then perhaps fail) or are acquired by other state firms. The OECD’s attempts at debt neutrality are inadequate because failure to repay debt has such different impacts on private firms and SOEs. The OECD’s calls for comparable return rates for state and private companies will not work: There is no way to make rates of return comparable when the two types of firms enter and exit in such different fashions.
Finally, should competitive neutrality be retained as an ideal goal? Yes, if the entities involved can be made fully cooperative. Recalcitrant governments have an enormous variety of tools available to them—regulatory and financial, formal and informal—to subsidize SOEs. It is not possible to anticipate all their actions. If such governments do not agree to truly comprehensive rules, implementation of competitive neutrality could be a game of whack-a-mole, with new principles rushed forward to address the ever-changing ways in which some governments evade the letter and spirit of existing guidelines.
What to Do
It would be a gaping weakness if the TPP and future agreements did not effectively check SOEs—the regulatory protection they receive, the financial subsidies, and their own non-commercial behavior. It is the correct goal to minimize the harm caused by their existence and behavior and, in that light, the ideal solution is simple: SOEs should be banned outright from as many economic activities as possible. The ban should expand gradually, allowing time for adjustment. It would not be complete, permitting a few SOEs to operate. This approach recognizes the powerful motivation governments have to create non-commercial corporations. A ban is hard to circumvent. Failure to move in this direction will harm the American and world economies.
However, banning SOEs from most sectors in many countries is an ambitious and difficult target. For the TPP, mass privatization is only feasible with a further, extended delay and considerable renegotiation, and perhaps not even then. In the short term, the U.S., Australia, and other TPP partners share the same goal: minimizing the pernicious effects of SOEs. The flaws of competitive neutrality as evident so far, especially regarding potentially uncooperative parties, are a major barrier. Detailed commitments must be won from governments that could behave strategically.
Because of options available to these governments, Australian and OECD notions of neutrality cannot be used in the TPP or most other contexts. Their flaws nonetheless do suggest some necessary components for competitive neutrality to fulfill the goal of minimal harm from SOEs:
- Differential treatment of SOEs cannot be justified by cost-benefit analysis. When SOEs inflict costs on private firms and individuals, these should always be rectified. The anti-competitive impact of compensation SOEs receive for carrying out public policy should always be rectified.
- Regulatory protection from competition is antithetical to neutrality and should be forbidden wherever SOEs are involved in commercial activity.
- The behavior of SOEs—their purchasing, sales, partnerships—should be purely commercial in nature and must not encourage national or local monopoly.
- SOEs should not be partially or fully exempt from World Trade Organization non-discrimination and national-treatment requirements. All firms should have the opportunity to compete for SOE sales and purchases. Government influence should not undermine these minimum standards.
- National, not local, governments should be responsible for documenting and ensuring compliance with competitive neutrality. The priority of creditors in receiving repayment should not be influenced at all by their status as public or private entities. Governments should have no preferential claim to SOE payments.
A secondary but still notable obstacle is lack of coherence in the U.S. position. High-level officials have endorsed “competitive neutrality” while failing to indicate which variant is being discussed. If the term is retained, it must be plain that what is meant is a powerful upgrade of the Australian and OECD versions both in objectives and implementation. The U.S. should put forward not just the best proposal of TPP members, but the strongest possible proposal that can adopted at present, addressing the full range of regulatory and financial advantages that can be conferred on SOEs.
The U.S. should therefore:
- Make clear within the U.S. government, in the TPP negotiations, and more broadly across international forums that old versions of competitive neutrality are unacceptable.
- Ensure the TPP variant of competitive neutrality can respond to governments strongly motivated to support SOEs (through regulatory protection, financial subsidies, and legalization of market-distorting behavior). The first of many needed steps is to strike the net benefit provision.
- Adopt an ultimate goal of barring SOEs from as many activities as possible. While not practical now, it may be feasible in, for instance, a future U.S.–EU agreement.
Conclusion: Or Else
SOEs undermine competition both at home and globally, and Congress has rightly shown bipartisan hostility to their current, prominent role. Unrestricted SOEs are a threat to an open global trading system. A weakly grounded approach to competitive neutrality will set a very poor precedent, sharply reduce the value of the TPP, and imperil its prospects for passage.—Derek Scissors, PhD, is Senior Research Fellow in Asia Economic Policy in the Asian Studies Center at The Heritage Foundation.