Saturday, May 8, 2010

Green Energy and the Learning Curve Myth

If you have been following discussions on "Green Energy", you have heard advocates tell you that the price of "Green Energy" will go down due to the "Learning Curve". I can tell you that 99% of these people have no real idea what the, so called, "Learning Curve" is. Try this question. Ask them, "Why is it called a "Curve"? and they will have no answer. If they answer that it is possible to "Fit" a second order exponential equation to most learning curves, you might be in trouble, but don't worry. It won't happen. The history of most "Learning Curves" is that every time the number of items produced is "doubled" the time to make one decreases by between 5% and 10%. Where the Myth comes in is that not all of the cost elements are affected by the, so called, "Learning Curve". The second error is determining the starting point for the number produced. Here is an example.

Ann is baking cakes and the cost of ingredients is $5 and it takes her one hour to bake the first cake. She feels that she is worth $10/hr, so she calculates that the total cost of the cake is $15. With me so far. Her son is in college and told her to use the "Learning Curve" to project her future cost of baking cakes. He tells her that she should use a 10% learning curve. 10% means that every time she doubles the number of cakes she makes, her time will go down by 10%. She now wants to figure out how much the 128th cake will cost. She is amazed that the 128th cake will cost her $9.78. This 128th cake will only cost about 2/3 of her "first" cake.

Here is the problem with her calculations. She has already baked a lot of cakes and is already getting good at it. Let's assume that she has already baked 64 cakes before she decided to try to sell cakes. Now when she uses the 10% "Learning Curve" that her smart college son told her about she finds that the 128th cake will cost her $14 and not the $9.78 she thought. This is where wind power gets off the track. The technology of building wind mills is over 500 years old and there have already been millions built. Most of the wind mill is made of raw materials that have been produced for the last 1000 years. These items are all subject to efficiencies due to a "Learning Curve", but to get a 10% reduction in cost we are going to have to make many millions more of them.

If you take something like the transistor, it did not exist before about 1945, nor did most of the technology to make them. We have thus seen a very significant reduction in their cost. When the "Learning Curve" efficiencies are combined with continual technological performance increases, the net cost has gone down even faster. The problem with all this "Green Energy" stuff is that it is old technology and thus requires very large volumes to make a significant difference in cost. The Lithium Ion battery was invented in 1904. It is not really new technology like the transistor. If you are following this, I think you can see that someone is trying to pull the wool over our eyes.

Take the automobile. Has the "Learning Curve" made the cost of cars go down? Look at the numbers. In 1950 you could buy a basic Chevrolet for $2,000 and now you can buy the same car for about $30,000. Back in 1950 you made $1.50/hr or $3,000/yr. Now 60 years later you make $80,000. Back in 1950 it took 66.7% of your annual pay to buy a car. Now it only costs you 37.5% of your annual pay. Conclusion, the "Learning Curve" on cars has caused the affective cost to go "down". Right? No! You forget one key factor. You are not back in 1950 with only the skill and earning power that you had then. You are smarter and more experienced and thus worth more. Think about how much you would make today if you were doing the same job that you were doing back in 1950. That job would pay about $45,000 today. So the new car would still cost you 66.7% of your annual pay. People forget that they have not been standing still over time while other things change around them.

Right now the cost of electricity from a wind mill has a cost that is almost 10 times that from coal. Based on the above, do you really think that wind power will ever be economical? Think about it. Even when the cost of coal power goes up, there will a be a positive feed back of cost into the cost of wind power because of all the inputs required to build and maintain the wind mills due to the cost of coal and oil energy. You can't hold one element constant over time and just consider what happens to your favored element. It does not work that way. As the cost of coal powered energy goes up, so does the cost of wind power and solar. The gap might change a little, but not by much. This is the "Myth" and it seems to be going down with the cool aid.