State versus Private Ownership

John Key's proposed running in 2011 on partial privatization of state owned assets.

The response from both sides has been pretty disappointing.

On the one side, besides the usual knee-jerk opposition to any kind of privatization and fearmongering about that foreigners might buy shares, there's the claim that we lose money by selling an asset that currently pays the government a dividend higher than the government's net borrowing costs. So if some SOE pays a 7% dividend to the government and the government's cost of borrowing is 5%, they reckon it makes more sense to keep the asset and to borrow money to cover shortfalls.

Forget SOEs for the moment. If any firm is providing a rate of return that seems to consistently be beating the market, we'd expect the stock price to rise until the rate of return falls into line with market norms, right? And if that doesn't happen, it's probably because there's something a bit nasty hiding in the risk profile. Now think about the SOEs. If they're earning a high return, it's either because their valuation is out of whack or because there's some risk. In the former case, the government can do well through an IPO - they'll get more for it than they thought it was worth. If instead it's just that the assets are risky, looking at the gap between funding costs and rate of return misses something a bit important.

Now, a reasonable counterargument is that the stock market rate of return is higher than the government's borrowing costs in general, so the asset price won't be bid up sufficiently to make the difference. But note two big problems. Sovereign debt from reasonable countries is safer than most stock market investments: the market index has to pay investors for the additional risk they take on. So selling a very safe asset (a bond) at a low interest rate while buying a riskier one (keeping an SOE) isn't a "Hey! Free Money!" deal. If it were, we'd also have proven that the government should borrow heavily on the international markets and buy up shares on the NZX. Most of us don't think that would work. So why do we think there's anything particularly special about the government's current set of asset holdings? If the argument for keeping Solid Energy in government hands is that the government's rate of return on coal investments is higher than its borrowing charges, then the government should also buy up any other firm providing a high enough expected return.

On the other side, folks largely overestimate the benefits of partial privatization. Sure, having shares trade on NZX is nice, but the government maintains a 51% share. There's no potential for an external shareholder to force changes in management if things are run inefficiently. We get some extra discipline from constant daily signals of what the market is saying about the firm's performance via the stock price, and if the share price plummeted, the Minister might want to have a chat with the CEO.

AntiDismal and Roger Kerr pointed out the limitations of partial privatization. Kerr worries that political incentives continue to be given too strong of weight in a partially privatized firm. Imagine for the moment that some town owned most of the local port company through a holdings company and that lots of retirees had put their money into this safe utility. At the margin, Council might prefer that money be paid as dividends to help keep local body rates down and to keep local retirees happy. Both make Council more likely to be re-elected. Problems stemming from deferred maintenance - those don't show up 'till somebody else is the mayor, and might be covered by insurance if there's a handy earthquake. Some folks arguing for partial privatization have pushed for restrictions on international share purchasing: the greater the requirements for local ownership, the more SOEs start looking like the stories about this hypothetical port and town.

I worry too that partial government ownership makes bailouts or other government support more likely. Socialisation of downside risk and privatisation of returns isn't a particularly good model, but incentives under partial privatisation lean that way at the margin.

Partial privatization seems unlikely to be worse than the status quo - it just seems insufficiently better to be worth the hassle. If Key's going to take flack for any use of the P-word, it would have been nice if he'd have gone just a bit farther with it. Here's a model I think could have worked well. Issue just over eight million shares in, say, Solid Energy. Just under half get sold via a float on the NZX. The rest are distributed, one each, to the just over four million New Zealanders. The SOEs get the benefits of full privatization. If "we" own the SOEs through the government, why not just hand us each a share and stop having the government as intermediary? The only plausible argument is that certain social goals are better advanced through state ownership than regulation. That's potentially plausible in regulation of natural monopolies, but we'd still need that the losses on the social side outweigh the usual benefits of private over state ownership. And while you could make that argument for the lines companies, you'd have a harder time doing it for a coal company

Andrei Shleifer noted the conditions under which we prefer state to private ownership. It makes little sense that we're privatizing our prisons before things like property valuation, red meat inspection services, or a coal mining company.