Alaska’s Megaprojects and the Sunk Cost Fallacy


Charles Wohlforth had a nice opinion piece in the Alaska Dispatch News recently, discussing one of Alaska’s failed megaprojects, the Point MacKenzie Rail Extension Project. At this point, Wohlforth shows, it is a hideously expensive bike path.

WC Is Always Willing to Help Republicans Better Grasp Reality

WC Is Always Willing to Help Republicans Better Grasp Reality

But it offers WC an irresistible opportunity to provide the Republican Majority caucus of the Alaska Legislature with an object lesson in the Sunk Cost Fallacy. WC has written before about the Sunk Cost Fallacy. At its simplest, this microeconomics principle states that you should not decide whether to spend more money based on the amount of money you have already irretrievably spent.

A “sunk cost” is an expense you have incurred and cannot recover. It’s never a justification for spending more money. There may be other reasons to spend more, but a sunk cost should never be a factor in the decision. The principle extends beyond spending more money; microeconomics states it should never be a factor in any decision.

Let’s use the Point MacKenzie Rail Extension to see how this works. As Mr. Wohlforth points out, the State of Alaska has already spent about $184 million building a railroad spur from Houston, Alaska to Point MacKenzie, on the north shore of Cook Inlet. The State of Alaska estimates it will take $120 million more to complete the spur. The spur is useless – except as a bicycle path – until it is completed. Should you spend the $120 million?

Well, you say, the State no longer has $120 million laying around, so the issue doesn’t arise. Perfectly true, but let’s pretend that oil is still at $100/barrel and that Governor Sean “Captain Zero” Parnell was reelected. Assume we have the money. Do we spend it? Not unless the project makes economic sense. The fact the State has already spent $184 million is completely irrelevant to the question of whether to spend more.

It’s hard to grasp. No one likes to waste money and it feels like you are wasting those first $184 million. Let’s try a more accessible example.

Suppose you impulsively bought a ticket to a Chicago Cubs baseball game. You can then choose between the following two end results if you realize that you don’t like the game:

  1. Having paid the price of the ticket and having suffered watching a game that you does not want to see, or;
  2. Having paid the price of the ticket and having used the time to do something more fun.

In either case, you have paid the price of the ticket so that part of the decision no longer affects the future. It’s the common factor to the two alternatives. You’ve spent the money no matter which choice you make. If the you regret buying the ticket, your decision should be based on whether you want to see the Cubs lose at all, regardless of the price, just as if he were to go to a free Cubs’ game.1 An economist will suggest that, since the second option involves suffering in only one way (wasted money), while the first involves suffering in two (wasted money plus wasted time), option two is obviously preferable.

It’s admittedly a bit counter-intuitive. It feels like you are wasting perfectly good money. When, in fact, you have already wasted perfectly good resources and shouldn’t be wasting more. There are elements of both loss aversion and overly optimistic probability bias, both of which incline you to commit the fallacy.

So, returning to the Point MacKenzie Rail Spur, the decision whether to spend $120 million turns on whether the Point MacKenzie Rail Spur is likely to be profitable, a source of jobs, encourage the development of Railbelt resources or provide some other tangible, demonstrated benefit. Not the $184 million that has already been wasted spent. The original concept was based on a study (PDF) by Dr. Paul Metz. Even Dr. Metz admitted the original projected capital cost would never be recovered by lower freight charges than traveling another 26 miles to the Port of Anchorage. Let’s be clear about this: the project, as originally conceived, couldn’t compete with the existing Port of Anchorage.

But the politically powerful Mat-Su legislative delegation, and especially then-Rep Bill Stoltze (R, Ignorance), the long-time House Finance Committee chair (and now state senator) was able to leverage the first $184 million exactly through creative use of the Sunk Cost Fallacy. Rep. Stoltze was unable to persuade his colleagues to the fund the whole original cost of $272.5 million in 2008. So instead, he persuaded the Legislature to appropriate $10 million in 2008. And then, in 2009, to spend another $17.5 million. And by then, Rep. Stoltze agued, they’d spent $27.5 million and it would be “wasted” if the Legislature didn’t carry on. So they did, with with seven more appropriations of cash and $30 million in state general obligation bonds. At any point, the Legislature could have said, “Whoa, this doesn’t make economic sense.” But no, that would be “wasting” the money already spent. The only reason more money was spent in any year was by commission of the Sunk Cost Fallacy.

WC recognizes that when trying to teach microeconomics to the Mat-Su legislative delegation WC is running smack dab into the Dunning-Kruger Effect. Rep. Stoltze doubtlessly sees himself as a hero for bringing home the bacon to his House district. An informed citizen would see Rep. Stoltze as pouring the state’s precious cash down a rat hole. As the Legislature struggles to balance Alaska’s budget, the expenses they cannot avoid include debt service on the $30 million in general obligation bonds that were spent on the project. It’s an annual reminder of the hard truth of the Sunk Cost Fallacy.

WC offers this lesson gratis; the State’s broke anyway.


  1. Just to be clear: WC expects watching the Cubs to be more pleasant this year than in some earlier years decades centuries. 
Advertisements