Friday, October 22, 2004

Energy independence

James Zogby is talking on the commonwealth club right now on KQED.
He made two claims:

One, that although alternate supplies of energy are good, we aren't going to achieve energy independence. This is probably true. But I think that replacing any substantial amount of our current imported energy with new domestic sources can help our national security and economic situation.

Two, that even if the U.S. imported no energy that we still would not be independent of Middle East oil, because other countries, on which we depend, depend on Mideast oil. Like Japan, China, Taiwan, etc. This may be true. But it's also true that if the U.S. develops a domestic energy production capacity, other nations will follow to the extent that it makes economic sense for them to do so. The U.S. isn't going to have a broadscale switch to domestic energy supplies unless those supplies are cheaper than the alternative. Oil was one of the first big Globalized products, so if it's cheaper for us it'll be cheaper for them.

Google at $400, part II

Is the Yellow Pages really a profitable opportunity for Google?


  • Google would be competing with a solution that really works already.


  • Local businesses are not going to abandon the Yellow Pages, so a Google sale requires them to purchase two listings -- more money in a down economy.


  • Where is Google going to list these ads? Am I going to use Google's website to search through JUST ADS? (That is what I do with the Yellow Pages right now, after all.) What criterion would Google use to rank the ads presented to me? If it's not money, why would the advertiser pay more?
  • Google at $400?

    Google, Yahoo, and Ebay are trading at P/Es of 207, 92, and 90.8. As the companies get bigger earnings growth will slow. Eventually the P/Es will come down.

    But let's think short-term: how about next March?

  • The companies aren't going to grow enough in that time to saturate any market. No need for P/E to change because of that.

  • Maybe oil prices, the crashing economy, and a new president will make the market more reasonable. Or maybe enough people will sell their falling retail stocks and head for the only bright stars on the horizon -- our three horsemen. Note that all three are already diversified out of the U.S. economy.


  • So a P/E of, say, 80 is possible next March.

    But what about E?



    Text ads are a better business than graphical ads. Nevermind that web surfers prefer text ads. They are better from the ad buyer's point of view, too.

  • graphical ads take a lot of manpower and often external contractors to lay out, which takes time and money.

  • that initial investment in layout is not closely predictable.

  • graphical ads, especially the animated kind with Flash, take more IT work to get onto a website.


  • So essentially, there is a bigger hurdle to cross if you want to use graphical ads. From google's point of view, the nice thing about text ads is that basically the purchaser's entire ad budget goes to google. It's also nice that an explosion of ads does not require an explosion
    of media contractors creating those ads.

    Right now, most businesses don't know much about how their online ads are working. Google's term sheet specifically prohibits advertisers from discussing with one another how well the ad campaigns are going. What buyers do know is that online ads have buzz, and Christmas is upon us.

    After the Christmas rush (and Google's year-end earnings release, which is going to be wonderful again), ad buyers are going to analyze what just happened. My guess is that three things are going to happen at this point:

  • online ads will turn out to be not as good as hoped.

  • Yahoo will be selling context sensitive text ads.

  • Major media companies who publish online content will also be selling context sensitive text ads directly, licensing the context bit from someone else (like Yahoo).


  • At this point, Google will be competing by saying: we can get more clickthroughs. They may also say they can get higher quality leads or broader coverage, but the measurable bit will be clickthroughs. This number gets put in the face of the guy who signs the check, it doesn't take any IT work to get it.

    So I see four downsides to the E part of P/E:

  • Google will face competition.

  • Major media companies will negotiate better terms to keep more of the advertising revenue already going through their web pages.

  • Buyers will be willing to pay less for a clickthrough.

  • The world economy will be in the toilet next year.


  • And one more thing: after Christmas there is a lot less advertising for a while. Ebay looks great as people shuffle presents to those who want them more, but retailers don't spend as much on ads.

    I suspect that momentum buyers, used to seeing 18% per quarter growth rates, are going to balk when they see a flat quarter. They're going to balk even more when all those unlocked shares start to sell. The fund managers who feel so left out right now will keep the bottom from falling out.

    Thursday, October 07, 2004

    Definition of Insanity?

    I've heard two different commentators on the radio recently state that a definition of insanity is doing the same thing over and over and expecting different things to happen.

    When I hand-saw through a piece of wood, I perform the same action over and over. For a while, it doesn't look like much is happening. Eventually, I cut through the piece.

    I shorted a stock. It kept rising, I kept shorting. Was this insane? Eventually it went down like I thought it should.

    The underlying problem here is that the things we act on in the real world often have state that isn't immediately apparent. We know it's there, we know we're acting on it, and we have to gauge the effects of our actions on that unknown state. Working with the unseen is common.

    Monday, March 22, 2004

    Nuclear Power

    I think we need a massive increase in the domestic production of nuclear power.

    Oil and natural gas and coal and electricity are, to some extent, fungible. If we produce more electricity with domestic nuclear, we can import less oil. There is a short-term limit, of course, because there are only so many oil-burning powerplants that can be supplanted by nuclear plants. But if new supplies of electricity make electricity cheaper relative to oil, people will switch other things from oil to electricity -- things like cars (plug-in hybrids) and home heating (heat pumps instead of furnaces and oil burners).

    Think of the advantages:

  • Oil is a major portion of our balance of trade deficit, so reducing our oil imports will improve our overall trade deficit. Our trade deficit is scary, because we pay for all these foreign consumables with domestic hard assets -- we are trading our land and buildings away for things that are thrown away in 15 years.

    Numbers: 9,140,000 barrels a day at $22/barrel (the bottom of OPEC price range) is $73 billion a year. The real number is likely a fair bit higher than that. The entire 2003 trade deficit was $374 billion, so oil imports were at least 20% of that.


  • The money we pay for oil goes largely to corrupt governments in unstable parts of the world (and Norway and the UK :). If we buy less oil, less money will go to these governments. Other nations will certainly step in and buy more barrels of oil, of course, but if total demand is reduced then the total revenue will go down with it. As the U.S. is the world's largest consumer of oil, by far, a significant drop in U.S. oil imports will have a very real effect on world oil prices and revenue.


  • Those governments spend their massive oil wealth in ways that are not aligned with U.S. interests. Saudi Arabia funds schools which teach millions of children religious intolerance and the glory of death in battle. Iraq was funding a militaristic dictatorship. Reducing the money they have to spend is in our interests.


  • External oil supplies are a national security problem. Right now, the world's producers are pumping at close to capacity. There is not a great deal of slack in the system. That means that a single large country, like Saudi Arabia, halting exports of oil would make for a massive disruption of the world economy. This gives the oil-producing countries a large amount of negotiating power. That their own economies would be wrecked by such a move tempers that power.

    But consider what would happen if the guy on the Saudi side of the table didn't care about the local Saudi economy. What if his priorities were religious and/or political?


  • Not just the rate but the total amount of oil is a long-term national security problem. The U.S. will use up its domestic reserves before foreign reserves are depleted. As U.S. domestic reserves are depleted, our reliance on and transfer of money to foreign sources will increase, and our independence of action in the world will decrease.


  • Shifting $73 billion a year into jobs which are almost entirely all in the U.S. directly employs about 730,000 U.S. residents, permanently. These people then consume services from others, and so there is some multiplier for the total number of jobs added to the U.S. economy. Though this is not going to solve our economic problems, it is enough to make a noticeable change in the unemployment rate.

    There is some possibility the jobs could be moved to Canada or Mexico. Either one of those two countries could invest in large nuclear programs and become a major exporter of electricity to the U.S. Canada is already a major exporter of hydroelectricity.


  • The U.S. accounts for a healthy chunk of the world's CO2 output. A significant cut in our CO2 production would help towards reducing the rate of CO2 rise in the atmosphere. The science on global warming is still out, but I think it's clear that humans are responsible for most of the rise in atmospheric CO2 in the last century, and I find it reasonable to imagine that that rise is changing something.


  • But of course the problem with nuclear power is the waste and the security issues. These issues look unacceptable if you think that the alternative is to simply reject nuclear power. But that's not the alternative. The alternative is invading nasty foreign dictatorships because we can't afford oil supply instability. But worse, we have to keep those foreign nations stabilized over dozens of years to ensure oil stability. That effort claims the lives of our soldiers who will perish trying to suppress rebellions overseas against what those rebels (correctly) see as U.S. interference in their domestic politics.

    So, I think we need to examine two propositions: Do we keep garrisons overseas in unstable nations for the next 100 years, and lose many soldiers every year to insurrection, in order to secure our overseas oil supply, or do we replace that oil supply with a domestic nuclear infrastructure that generates, uses, and discards enormously dangerous substances as part of its basic operation?

    And my answer is, better the devil you know that the one you don't. Nuclear is an entirely domestic industry. Our government has the ability to regulate this industry. The regulation may not be perfect, but at least everyone involved is inside our borders. Nobody questions that the U.S. military and police forces have the right to secure nuclear facilities. There are no protests in the street from those willing to die to prevent the NRC from specifying standards to which the nuclear industry is accountable.

    Of course, I would like to see some changes in the nuclear industry. Reactors are still built as if a major goal was the production of weapons-grade plutonium. The only way to reduce the amount of nuclear waste that must be separated from the environment forever is to reduce the interaction of radioactive stuff with the environment. This means that reactor cores should be sealed objects, built in the very crypts they will stay in for hundreds of thousands of years. The only thing that ever comes out is heat. When the fuel in the core is "burned up", the core is irreversibly shut down and left to cool, essentially forever. New cores, with their own independent containment vessels, are built next to the old ones, and the power generation infrastructure is switched to these new cores as the old ones die.