After reading a paper titled The Railroads and the Raj, by Dave Donaldson, and the ongoing controversy about building out a high-speed rail line between Los Angeles and San Francisco/San Jose, I’ve been thinking a bit about how to fund projects where the payout is over a very long time horizon. I haven’t gotten far with it, so this post is mostly thinking out loud.
First of all, the average business investment is funded out of basically two sources: the business’s private savings (or the savings of the owner, in a privately owned firm), by issuing shares of stock to would-be owners, or by borrowing money from banks or other people holding savings. If you’re thinking about opening a food truck, you need to either buy or lease a truck. Let’s say you buy it, but you don’t have enough cash: then you need to borrow a few tens of thousands of dollars to purchase it. In exchange, you tell the bank that you will give them some cash next year. To convince them, you have to tell them how you plan to get this cash: by making gourmet grilled cheese sandwiches and selling them for $5 to people outside of bars late at night. If the bank agrees with your revenue predictions (or if they have faith in your ability to find a new bank next year to borrow from, and so be able to pay them back from the new borrowing), then they lend you the money. If you don’t make enough money over the next year to pay them back, but you make some, then next year you borrow a smaller amount to repay the first year’s loans. At this rate, it takes you a few years to pay them back, but if you have some success than it works.
Hooray! This is the magic of the banking system: if you didn’t have access to the banking system, you wouldn’t have started the truck, and I wouldn’t be eating your delicious grilled cheese twice a week. It would be a sad world.
Now, some projects can’t be repaid in a year or two, no matter how rosy of projections you make. If you’re building a car factory, or an iPhone factory (which you aren’t, at least not in this country), the cost of construction might be so high that it takes five or more years in order for the profits to repay. These projects face two distinct challenges: (1) the total amount that they want to borrow is going to be very large, and (2) the time horizon for repayment is fairly long, and banks might not be willing to lend for that long. However, since firms needing to build factories (or other similar projects) are usually established firms–think Google building a massive data center, for instance–they don’t face the same constraints that you do with your food truck. Google sold shares of ownership to the general public starting in 2004, and people were quite willing to give Google some cash in exchange for a share of future profits. This enabled Google to acquire billions of dollars in financing–plenty enough to fund data centers, solar power centers, small firm acquisition, and anything else they’re likely to need to fund. And at that point, if they want to, Google has plenty enough collateral to convince banks they are a good risk, so Google is actually able to borrow if they want to. This process is actually pretty sweet: firms start out risking small sums of money, and successful ones (that is, well-run firms with good business models) are able to raise more money to pursue their proven ideas.
So far so good! We have a system that enables a wide variety of firms to access funding in various ways, and it has a natural selection process by which firms with better business plans can access more money, and those with risky business plans have to prove themselves before they gain access to huge amounts of funding.* However, there are still more possible projects that don’t fit these outlined parameters: those that are so large, and/or so long-lasting that the firm may take decades to produce enough profits from the project to repay the borrowed amount. A few examples that I have in mind are railroad networks, power stations, and oil refineries. The open question is, how best to fund these projects? Because of the basic facts that (1) the world economy tends to grow and (2) people buy more oil as they get richer, oil refineries tend to get funded okay. But even so, basically zero new refineries have been opened in the US in decades. Decades!
Trains are a more interesting problem to me. They are competing in the transportation sector with (ostensibly) private firms that sell cars and airline tickets. The catastrophic bankruptcies of 2/3 of US auto manufacturers and the semi-perpetual state of bankruptcy endemic to the airline industry notwithstanding, many are convinced that building a new high speed rail network is a hugely expensive and risky proposition. And well, they are correct–it is hugely expensive and a risky proposition! It’s a huge upfront capital cost on an unproven concept in a country built around cars and planes. However, with a little luck on the front of immigration reform (luck that seems possible–the 1990s and 2000s were a golden age of immigration in this country), the US will be home to 500 million people within my lifetime (I suppose also a little luck on the longevity front). That is the current population, plus most of it again. If it were spread evenly throughout the country, my adopted home metro area of LA will have a population of almost 30 million people. The Bay Area would be almost as big as LA is now.
However, the newcomers will not be spread evenly throughout the country. They will make sensible choices about which cities to live in. If they continue the trends of the last 60 years, then that will mean that our suburbs will expand to the point that we will have not just a Boston-to-Washington corridor, but a Boston-to-Washington-to-Atlanta-to-Chicago-to-LA-via-Seattle corridor. Traffic will suck.
The alternative to adopting all this new car-based infrastructure, is to build more densely in the cities we currently have, and build alternative infrastructure to enable people to move between cities–perhaps like trains! But that might not happen, and the alternative seems to be an incremental increase in suburbia until we have to rechristen ourselves the United Suburbs of America. Which makes the basic question here important: how can we manage to fund long-term, risky investment projects like this? It’s an open question, and a seemingly difficult one, given the difficulty that even the wizards of Wall Street have had in answering it.
My proposal: borrow money from China. It seems to work.
*Yes, I’m aware it doesn’t always work out this way in practice, but I’m focused on a different problem today.