Investment, Dynamic Consistency and the Sectoral Regulator’s Objective

The opinions expressed in this article reflect only the authors’ views, and in no way bind the institutions to which they are affiliated

In some countries, regulators are appointed for longer periods than legislators. However, laws are usually the result of a complex political process and usually much harder to change than sectoral rules and regulations. Take, as an example, the longevity of the Telecommunications Act of 1996 in the USA. Another example of longevity is the Regulatory framework in the EU applicable to the electronic communications networks and services, known as the 99 Revision of the European Commission, which was enacted in 2003 and is still in place.

See Guthrie (2006) for a discussion about the regulators’ inability to commit to a regulatory policy.

This is related to the literature on strategic delegation, namely in the context of the relationship between the owner and the manager of a firm. See, for instance, Fershtman and Judd (1987, 2006) and Sklivas (1987) who argue that it may be profit maximizing for an owner of a firm to distort the manager’s objectives away from profit maximization.

Other studies, which convey the idea of the existence of an optimal bias between the politicians’ and the regulator’s interests due to information asymmetries, include De Figueiredo, Spiller, and Urbiztondo (1999), Sloof (2000) and Spiller (1990).

There is also some empirical literature on the link between the regulator’s preferences and investment, which has produced mixed results. Trillas and Staffiero (2007) point to evidence that in many developing countries, and especially in Latin America, some degree of industry orientation has been necessary to attract foreign capital in the utilities sector. On the other hand, Gutiérrez (2003) investigates the relationship between the regulator’s independence from the industry and from the government and investment in telecommunications infrastructure, finding that greater independence positively affects the number of phone lines per capita.

An alternative to changing the regulator’s objective function is to constrain its future actions. The legislation behind the Kansas Telecommunications Agreement prohibits the regulator from carrying out earnings audits, thus making it harder for the regulator to review a price cap (see Weisman 2002).

Strausz (2011) shows that regulatory risk, reflecting the uncertainty behind new or changing regulation over time, may actually benefit firms when aggregate consumer demand is convex.

This solution, which follows the proposal by Rogoff (1985), is a second-best commitment mechanism, where the pricing decision is delegated to an independent regulator, whose preferences do not necessarily coincide with those of the government. Levine, Stern, and Trillas (2005) make an analogy of the Rogoff-delegation solution for central banks to other regulatory environments.

We model the investment decision as a discrete choice since investment in many of these industries has a discrete nature. Take the example of the deployment of a next generation network. It involves two aspects: upgrading the core network and replacing the local access network. Upgrading the core network is clearly a discrete decision. Replacing the local access network amounts essentially to replacing copper lines by fiber optical lines, with the associated electronic control equipment, etc. Again this is a discrete decision. Furthermore, there are estimates that 68% of the costs of this investment relate to the construction of ducts. This means that most of the costs are fixed and irreversible and do not depend on the type of equipment installed.

Given our assumptions about payoffs on Section 3.5, there is no need to give a weight higher than 1 to the entrant’s profit. The legislator can obtain a similar effect by adjusting the weights to the other parties’ payoffs.

Note that in principle one should have , since the higher the quality of the network the entrant is using, the higher must the access price be for the entrant to prefer to stay out of the market.

See DeGraba (2003) for a derivation of the relationship between the access price and the incumbent’s price.

We assume that there is no price-squeeze regulation in place that would force the incumbent to charge a retail price above the access price.

We assume that the legislator provides the regulator with a tie-breaking rule, such that the optimal access price is unique, e.g., choose the lowest value or the highest value.

It is also possible that , in which case a higher weight to consumer surplus would make the regulator’s optimal access price equal to the first best.

Suppose one assumed instead that the incumbent’s profit is increasing in the access price until a given , and decreasing afterward, i.e., the profit maximizing access price leads to duopoly. Then, for a sufficiently high weight to the incumbent’s profit the regulator would set , and the entrant would enter the market. Our results would continue to hold qualitatively.

Note that if constraint [3] is binding, i.e., it follows from Lemma 2 that is increasing in k. Otherwise, is independent of k. Moreover, increasing k reduces the set of values for that verify constraint [3], thus is non-increasing in k.

In case a bypass investments were possible, i.e., the entrant could invest after the incumbent, the optimal policy would involve the legislator setting weights conditional also on the entrant’s investment decision. If, for instance, a bypass investment were socially desirable, the legislator might set a policy such that the weight given to the entrant’s profit in case of no investment by the entrant was lower. Again, the solution would be such that the regulator would choose an access price after investment equal to the one a regulator able to commit would choose.

The European Commission has been working on a regulatory approach intended to promote advanced access infrastructure investment in the telecommunications sector. As expected, incumbents are in favor of high access prices, whereas entrants argue for low access prices. In 2011, the European Competitive Telecommunication Association, ECTA, based on WIK (2011), suggested that the incumbents charge high access prices in the cooper network, with the excess profits reverting to the entrants in the absence of investment in fiber. This suggestion was eventually turned down. We thank one referee for pointing out this example.

Note that a pro-incumbent regime is characterized by the regulator placing a weight on incumbent’s profits that is larger than the weight placed on the other parties. It is not necessary that value is explicitly stated in the regulator’s objective function. In fact, could be the focal point of the set of weights larger than one that verify some additional restrictions that constrain the regulator’s actions.

As corresponds to the increment in the incumbent’s gross profit when the regulator’s objective function is not distorted, its value is the same both when the weights are and are not conditional on the investment decision. The same applies to .

The determination of is similar to the case where the weights are conditional on investment with a difference in the incumbent’s constraint [3], which is now harder to verify as the regulator is unable to punish the incumbent in the case of no investment.

The incremental profit of investment is maximized at Therefore, it is not possible to induce investment if k is on