S Aufrecht

Liveblogging my econ class just for the heck of it.


  • The difference between capital requirements and reserve requirements for banks. The current reserve requirement for US banks is that they hold cash in their vault (or on deposit with a Federal Reserve Bank) equal to 10% of outstanding deposits. Hmm, does each reserve bank in turn need to keep cash equal to 100% of all of those deposits on hand, or do they also get to do reserves?
  • Aggregate supply and demand: Macroeconomists got jealous of all of those supply and demand curves that microeconomists get to draw so they made their own.
  • Phillips curve. Apparently inflation correlates inversely with unemployment. But the model doesn't hold up with more data (although Donald Luskin may not be a reliable source, even for graphs. There are various explanations about how all of this works, like employees adjusting to rising inflation and building inflation predictions into their wage expectations. But they all strike me as Just So stories. Everybody has a predictive model with a prediction horizon o negative one: the models each predict the last thing to happen. When the model fails to predict the future, a new model is created to explain the new factors that made the first model fail.
  • There is no relationship between economic growth and wealth inequality in the United States in the 20th Century; see graph.
  • The current treasury yield curve, which is apparently in a normal state. As per the previous point, "There are four main economic theories attempting to explain how yields vary with maturity."
  • Deregulation in the US. Here's a nice graph of how we've gone from one AT&T to about fourteen baby bells and then back to three (AT&T, Verizon, Qwest)
  • The Fed chairman is given independence from politics. hah!
  • I do have to say that the topic, if not always the substance, of macroeconomics fascinates me. Millions of millions of people making complex decisions and one can almost, almost, describe and predict their behavior in the aggregate. The notion of tracing bundles of greenbacks from the Fed to banks to getting spent five or ten times per year, perhaps in other countries; of the whole fiscal merry-go-round that may or may not be a Ponzi scheme. Maybe it's just money, not macroeconomics, that is so sexy.
  • I wish this class were less US-centric. Perhaps that's because most economics work in and study the world's biggest economy? But the US economy isn't the majority of the world economy.
  • The main goal of monetary policy is price stability. Singapore's Monetary Policy


  • Some topics for next term: monopolies
  • Guns, Germs, and Steel. I went digging for my review only to realize I wasn't blogging in 1997/1998. And my email archives are off-line before 2001. Nuts. Interesting to see the criticism that's accumulated, some of which rings true just in the Wikipedia summary
  • Costs. Opportunity cost, sunk cost, economies of scale, economies of scope. Apparently I'm about a month ahead of our reading, which makes me feel better about falling behind the syllabus lately.
  • I think the key insight between many of these cost ideas is optimizing for context instead of applying a global optimization rule. And lo, the title of the first slide is "Principle: Relevance."
  • Economic Value Added = Revenue - all costs, most notably cost of capital. If you can turn $100 into $120 in one year, you probably did well. If you can turn $100 into $120 in ten years, you did poorly. A charming story about how the EVA consultants helped Herman Miller unlock their potential in 1995-1997. Apparently it's a one-time benefit, though.
  • What's the opportunity cost to Singapore of having the Istana (the presidential palace) taking up prime property? What's the benefit? A classmate notes that one sector of Washington D.C. has not benefited in the recent improvements in D.C.'s conditions because of a homeless center. Should the homeless center be moved to cheaper land so that the land can be improved? (How many homeless are there in D.C., by the way? about a thousand? 15 to 40 thousand?
  • Example: if Apple bought hundreds of millions of 1-mbit DRAMs in the summer of 1988 for $38 each, and then the market priced dropped to $23 each in January 1989. For setting the price of Macs in 1989, what is the cost of a 1-mbit DRAM in inventory? (Contemporary Macs had 1 MB of RAM, so would need 8 chips. Unless Macs used parity bits, which I seem to remember they used to, in which case they'd need 9.) (This has sent me on a quest for historical DRAM price charts, which despite access to substantial commercial databases is straining my powers of search. Class continues through economies of scale and scope, which I believe I'm already on the record as supporting.)
  • Yes! Finally tried a Google image search on "dram price history", which led on page four to this, which mentioned a 1994 article in the Journal of Political Economy, which I was able to retrieve through EBSCO.
  • The cost of a pilot for a plane is a fixed cost—you need two pilots for each flight, regardless of size of plane or number of passengers. But it's not a sunk cost, because you can fire the pilots. Question: Is the depreciation of an airplane a sunk cost? Yes and no.