The million dollar pencil factory problem - Economics and Defense Contracting

 The million dollar pencil factory problem - Economics and Defense Contracting

A common economic problem, and one that is often not well understood, is the million dollar pencil factory problem. If a million dollar pencil factory makes only one pencil, that is effectively a million dollar pencil. In order to make up for the cost of building an entire factory, you have to make enough pencils to justify the initial cost. One might assume that one then needs to make a million pencils; however, that would be incorrect, as no-one buys pencils for a dollar. You wouldn't buy a pack of 50 pencils for 50 dollars, you'd buy it for 4.99, or 5 dollars. As a result, you need to manufacture at least 10 million pencils to make your money back. 

However, there's additional costs, such as labor for the workers, their training and pensions, electricity, and of course the biggest part of all, the actual cost of the pencil, the wood, the graphite, the metal and eraser. To break even just on the cost of the pencil factory, you need to manufacture, and find a way to sell, 100 million pencils. It's not enough to make them, you need a buyer. Advertising, sales teams, some kind of way to make your pencils worth it on the market. By the end of it all, you need to have manufactured an entire line of pencils. Then there's staying in business over many decades. 

The problem with a million dollar pencil factory is that if it only produces one pencil, that is a million dollar pencil. The cost for research and development, prototypes, design, sales, studies, marketing, and the like are all things that must be factored in to the price. Mass production, naturally, reduces prices, but not always in ways people think of. The same is true frequently with defense contractors. 

Companies which sell equipment to the military have it worse than normal, as they have a market of one, essentially. What that means is, unlike a civilian who can go to a store, pick up a can of beans off of a shelf, decide they don't like it and return it, there is no market for tanks outside of the military or collectors who have limited legal capacity to obtain them. There is little market for rockets outside of NASA and Elon musk. When you create a custom product from scratch, you need an entire factory line set up for it. There are not items you can just return, there are sunk costs for the base cost of developing something. Before a single finished F-35 had taken off of a runway, the U.S. had already spent 400 billion dollars on the program. In the program's lifespan, it will be worth approximately 1.6 trillion dollars for all 3000 F-35's. However, 1/4 of that was spent before a single finished plane took off the ground, just in research and development, creating factories, training workers and for prototypes. 

When Donald Trump sold F-35's to our allies, increasing the total number of aircraft, the cost per aircraft went from 240 million dollars per plane, to approximately 80 million. The primary reason being that the base cost of manufacturing, the factories, workers, electricity etc. and research and development was all fixed. The cost of going from 1500 aircraft to 3000 was only a marginal increase in the overall program's cost, from approximately 1.45 trillion dollars to 1.65 trillion, but resulted in twice as many aircraft. 

Often times in many people's minds, to cut spending they want to cut the number of something being made, but it doesn't work like this in production. By investing in a system and mass producing it, we can produce more of it, and lower the cost, within a certain extent. Before hitting diminishing returns, it behooves us to make an appropriate amount of something to get our moneys worth from a project. Simply going from 10 to 9 aircraft carriers might not save much money; in fact, it might not save money at all, as the factories that needed to be built to build that many aircraft carriers is where the bulk of the cost comes from. It's not some abstraction of steel and dollars, it's a real world project requiring a real port and factory to build. Research, training, technology gaps, solutions found, the sunk cost of which can be undone. 

One might believe it would be cheaper to make less aircraft carriers, but it could be cheaper to build more and sell them to our allies to spread out the base cost of the product. Rather than buying less than we need and having to buy more later of different types, buying a lot of a product that is meant to last well in to the future and that can be upgraded and consolidating around a single chassis or base unit will save the most money. Eventually when the factory workers move on to another project, they will not have to work about job security as they already built their old product and this won't be put in to question. 

Understanding the million dollar pencil factory problem is key to many misunderstandings in economics, something that can't be resolved by simple number crunching alone. It can be non-intuitive at first from an ordinary person's perspective of buying things, such as from a grocery store, as bulk purchases of custom products are generally uncommon. It's hard to imagine a need for buying 10,000 custom ordered birthday cakes. However, it's not as if those chefs can simply undo their labor. If you order this, you will have to pay for either way with no refunds. The same works for the military and defense, so simply cutting ships from 10 to 8 or tanks from 10,000 to 8,000 might not even save money at all, but rather just hurt our military readiness while politicians can gloat about "doing something". Fundamentally, it would be cheaper just to sell these to allies than to not make them. 


An example is the zumwalt destroyer, while originally 100 were to be made for 30 billion dollars, instead it transformed in to 30, then 10, and then 3, with congress not understanding why each ship now cost 7 billion a piece, instead of the original 3 billion. The simple reality was the cost to develop the capacity to produce 100 ships had a base cost of 21 billion dollars. Rather than declaring the zumwalt a failure and scrapping the program, they should have build 100 ships which we needed and gotten the costs down. But this kind of thinking results in tremendous waste, ironically something congress claimed it was trying to avoid. 

While especially true in defense contracting, it is also an issue in other areas. A famous golfer had a custom made golf club for himself, and sold thousands of them to his fans so he could justify the cost of making a single one. While obviously not custom fit to each fan, they could buy a similar enough product for themselves and make their favorite golfer capable of having a batch of very well made golf clubs. To make even one of these, he first had to get a factory to make 10,000. The same can be true of other products. Therefore rather than thinking you can simply cut a certain number in production, it is better to manufacture a certain fixed amount which results in the most economic efficiency. When the U.S. military developed over 500 tanks a year vs. 50, tanks were cheaper overall, as more of the cost for labor, the factory and other factors were spread out over each tank. Therefore it behooves us to consolidate around a single tank chassis and make more of them rather than make dozens of unrelated vehicles; using a tank body for an IFV, bridge layer and the like would save money on each tank by spreading out the load cost to more vehicles. While it may not be possible with the abrams inherent complexity, designs optimized for this way should be considered wherever possible. 

Thus one needs to take in to account what I have coined "the million dollar pencil factory problem" when considering defense contracting and various other means of mass production. The simple reality is that mass production, when done correctly, can lower prices, and both for military readiness and economic efficiency, it behooves us to take this in to account when creating contracts.