Thursday, October 22nd, 2009
This post kicks off a multi-part series on the principles of privatization. Cities and states across America have been looking at various types of privatization programs to save money or raise money in a tough economy. This has been surprisingly non-partisan, with proponents and opponents scattered across both parties. Democrat Mayor Richard Daley is a huge fan, as is Republican Gov. Mitch Daniels in Indiana. Democrat Gov. Ed Rendell pushed a toll road privatization. The Republican administration in the city of Indianapolis is looking at several privatizations, including airport parking garages, the water utility, and convention center operations.
Done right, privatization can yield big benefits. But done wrong, it can be bad thing. My own view is non-ideological in that while I see the virtues of privatization in many cases, I think we have to look at the actual transactions in question to see if they are any good from a practical perspective. Often they are not.
Part of the challenge is that there are so many different types of privatizations. The word “privatization” is often used to describe radically different types of things. Also, many other things which are de facto privatizations are implemented without using the term. And even within a given deal, many different types of transactions are, in practice, bundled together.
In this post, I wanted to try to sort these out, and talk about various types of transactions and what they are. It is important for anyone considering a privatization transaction to keep what it is we are actually proposing to do in mind. Otherwise we might draw false analogies with other transactions and end up making a mistake as to what we are actually doing. I think there are three broad categories of things to consider: basic transactions, policy changes, and financial engineering.
Throughout this series I will return to the lease of the Indiana Toll Road and Chicago Skyway as examples of good privatization transactions.
Defining what a privatization transaction is can be difficult. I’ll define it roughly as a transfer of government funds or assets to a private organization in return for a payment and/or the execution of a government function. Some of the major types of things that might fall under this or some expanded view of it are:
- Contracting. This is simple and well understood. The government hires a private company to provide a transactional service, for example, to build a road. This might not seem like privatization per se, but in theory governments could do everything in house. Indeed, they typically do perform routine maintenance and repairs with their own staff, such as pothole filling.
- Outsourcing. This is also a type of contracting, but involves delivering a service over a particular and often extended contract term rather than for a particular transaction. For example, companies often outsource their IT or routine business processes like invoice processing. There can be big value in this if done right, which I talk about in a future post.
- Asset Sales. A government could simply sell off assets. This is done both at the large scale, such as by selling off state owned industries, or the small, such a routine surplus property auctions. The federal government has acquired quite a large number of assets recently in the form of failed banks and stock in bailed out companies. The intent is that this be temporary and that these be sold at some point when conditions change.
- Operating Lease. This is where an asset is transferred to a private company for a period of time in return for periodic payments, and where the private company takes over responsibility for operations as in an outsourcing contract. The Indiana Toll Road lease falls into this category. You might also think of a concession, such as a lodge at a state park, as falling into this category.
- Sale-Leaseback. This is where the government sells an asset to a private party, then leases it back. Generally at the end of the lease, the government buys it back for a pre-negotiated price as well. You might be asking why anyone would do this. In the public sector, it is a manipulation of the tax code. Governments don’t pay income taxes, private companies do. So governments don’t get any value from depreciating an asset on the books. But if they sell it to a private company, that company can deduct the depreciation on its books. In effect, that’s the sole reason for doing this deal. The company and the government split the tax gains. This used to be more common. The Chicago Transit Authority did a sale-leaseback on its Green Line L this way, for example. However, the feds were understandably peeved by this practice and I believe have largely shut it down. In the private sector, sale-leaseback transactions are typically a form of financial engineering.
- Public-Private Partnerships – For Profit. “Public-private partnership” is another nebulous term, but in this case I am referring to publicly subsidized real estate projects. In effect, rather than contracting with a private company for a government service, the government dabbles in traditionally private sector areas like real estate. Generally in this case the government will do things like assemble land and put in extensive infrastructure, often using tax increment financing, and private developers put in money as well and also operate the project. Sometimes there is just a pure subsidy to the development, but not infrequently the government actually has an equity stake. I recently talked about how Kansas City just got a $1.8 million check for its interest in arena profits, for example. Government usually do this in order to promote development in areas where the market isn’t investing, such as downtowns or impoverished areas.
- Public-Private Partnerships – Non Profit. There’s a less well known side to public-private partnerships, one that flies under the radar and isn’t typically thought of this way, but is actually one of the oldest and most important privatization transactions. That is when the government gives grants to non-profit organizations to implement projects or run programs. These are often social service or arts related. And often the public money is pooled with privately raised funds. This one is interesting. Privatization is usually associated with the political right and criticized by the political left. This type of transaction, rarely considered “privatization” but which clearly is, is generally associated with the political left and criticized by the political right. The rationale for these programs is generally that the government can help multiply the effect of private donations, and also that these non-profit groups – such as public arts programs or community development corporations – are closer to the needs of particular communities and bring unique expertise to the table.
In addition to these basic transaction, privatization can also involve other public policy changes. There are three that I typically see:
- Elimination of Services. One way to look at privatization in its most pure form is if the government simply stopped providing a particular service and left it to the marketplace to provide, similar to how most products and services work. This rarely happens, but one example where you see wide variability is municipal garbage collection. In some cities and towns, the government provides residential garbage collection. In others it is does not. In those places, people have to contract with a private trash hauler for a fee. Large cities tend to provide residential garbage collection as a service, and with large numbers of very poor people often living there, abandoned properties, etc. stopping this might not be such a good idea. Smaller places with more uniform incomes can more easily pull it of.
- Addition of Services. The government could add a service such a building a new toll road or transit line.
- Pricing Changes. A price change could be involved. For example, most lump sum privatization deals like the toll road leases involve significant price increases. Sometimes cities implement fees for services like trash pickup to raise additional funds.
Lastly, there can be various types of financial engineering involved. By that I simply mean transactions that involve buying or selling financial instruments at market rates. Similar to buying or shorting a stock, the value in these can be positive or negative and is only revealed over time.
- Capitalization. This is where you take a stream of future revenues and convert it into a lump sum. It’s just like lottery winnings where you have the choice of a large amount of money paid in annual installments, or a smaller total amount of money all given to you at once. (In reverse, you borrow a lump sum now, such as your house price, and pay it back in installments over time in your mortgage payments). Is this a good deal? Well, a dollar today is worth more than a dollar tomorrow, something known as the “time value of money”. In order to get someone to exchange a dollar today for a dollar tomorrow, you have to pay them interest to compensate for this. Whether there is any value in this is a complex topic I won’t go into completely now. But one area where I would say Yes is if you are able to invest the lump sum proceeds into something that earns a greater return than the interest you paid on the transaction. For example, construction costs for highway and transit systems had until recently been increasing at a rate greater than inflation. Due to regulatory creep, this would appear to be the case for some time. If that construction inflation rate exceeds, for example, the discount rate on the projected revenue stream in a toll road lease (or the rate you would pay on bonds to borrow the money to build now), it can make a lot of sense. That’s one reason I think the Indiana Toll Road deal was very good. But taking a lump sum and investing it in new construction quickly, Indiana was able to avoid a large amount of project inflation over the decades it would have taken to build those projects using traditional means. Realistically, the state would never have been able to afford many of them.
- Hedging. A hedge is a type of financial transaction you enter into in order to reduce your exposure to risk. For a fee, some other party takes on that risk. This could be through various formal or implicit mechanisms, but it fundamentally a derivative transaction. If when you think derivatives, you think financial meltdown, you are right. Derivatives are often very risky investments, though for traditional purposes they can be very useful because they provide predictability. For example, airlines know that they need jet fuel every year in the future. Rather than simply pulling up to the pump and paying the cash rate of the day, they can buy futures that allow them to lock in future prices. That way they know exactly how much they will be paying next year and can plan accordingly. Of course, if you lock in now and the price goes down, you lose money on the deal. Or if you need less than you thought, that’s also bad news. Investing hedges is often pure speculation. Because of the risk of losses if things go bad, governments tends to avoid hedging because of the associated political risk. However, hedges are often an implicit part of privatization transactions and can be a source of value because of the so-called “Winner’s Curse“, where bidders often overpay in an auction.
Keeping in mind all of the different things above, when looking at a transaction it is very important to understand what is part of the deal. Often multiple of these transactions are combined.
Let’s examine the Indiana Toll Road lease transaction. This involved a bundle of many different types of transactions:
- A price increase via a significant increase in tolls
- An operating lease where the asset is turned over to a consortium for 75 years and the consortium is responsible for operations and maintenance
- Capitalization of the lease payments/tolls into a $3.9 billion lump sum
- A contract of $400 million to the consortium for capital investments in the Toll Road
- A hedge against future fluctuations in toll revenues, operations, maintenance, and expansion costs.
- Addition of services (electronic toll collection)
As you can see, this was in fact a very complicated transaction. The fact that it largely took place with a single bidder and a single contract can obscure this complexity. Indeed, behind the scenes I am sure that the consortium that signed the lease actually parceled out many of the transactions elsewhere – to banks, construction companies, etc.
Why package these together? Some have asked the question, at lease implicitly. One variation goes, “Why can’t the state just raise tolls itself?” The question almost answers itself. There is enormous pressure to set the price of user fees for government services very low. Raising fees, like raising taxes, is often very unpopular. That’s the reason the Indiana Toll Road was a money losing asset to the state. However, by packaging these together, that toll increase is magically transformed into $3.9 billion in the here and now. That provides a powerful incentive and political justification for the tax increase. Lawmakers can go home and explain all the projects that are going to get done with the money.
As I said, I think that the Indiana Toll Road lease was a great deal. Similarly the Skyway transaction. The city of Chicago converted a financial albatross into gold. However, other transactions that might seem superficially similar – such as other asset leases that involve lump sum payments, may not be nearly as good because they don’t contain the same mix of elements. We need to carefully examine the various components of the transaction and where the value is coming from. That’s the topic of the next installment.