Maybe we could start out with an 80% tax on profits from stocks sold within 3 days, and gradually increase it to 6 months?

I’d expect liquidity to go way down. Brokers would lay off staff.

Maybe the options markets could take over? When you buy a stock you sell the right to buy it in 6 months at some price. Maybe at 4 months the price for that option goes down enough that you buy it back, and then you sell it again at 3 months. Etc. Even though the sales of the stock are regulated, if sales of the options stay unregulated then people might manage their short-term bets that way.

Maybe you could sell an option to sell at market. Then someone who thinks he wants to hold the stock a shorter time could buy the option, say, a few days ahead of time. When it comes due if he wants to sell then you sell for him and he doesn’t pay the tax because you’ve held it for 6 months. If he wants to buy after all then he buys it from you and then he’s stuck with it for 6 months—but he can sell the option.

I expect the biggest result of a tax like that is that the finance industry would stage an all-out advertising and astro-turf-roots campaign to claim that the tax is anti-free-enterprise, anti-liberty, anti-industry, anti-capitalism, anti-American-Way, anti-prosperity. That it destroys americans’ retirements. That it empoverishes widows and orphans. That it results in a flight of capital from the USA, the investors we need to keep America great will all move all their money to foreign stock exchanges and US businesses will all go bankrupt and the Treasury will have to default on savings bonds.

There would probably be a move in Congress to impeach a president who let the IRS make a ruling like that. Surely Congress wouldn’t pass such such a tax law. It would be too unpopular.

Hmmmm … I would be almost prepared to entertain this in respect of the size and value anomalies, but it’s quite hard to see what sort of risk momentum anomalies would proxy for.

Look at the momentum chaser blowup in the hedge fund community in 2007!

Evidence is very ambiguous here as far as I’m concerned – it is not like “buy and hold” has had fantastic returns these last 12 months either, and the strategy outlined in Khandani & Lo’s “What Happened to the Quant Funds in August 2007?” actually made back most of its drawdown in the following months.

Bear with me a moment.

Suppose that John Quiggen is the brother-in-law of an executive secretary at a public company, and she has handled some very interesting documents that convince him this is the time to buy. And he tells me. I trust John, so I buy. But I can only afford to risk $10,000 on this inside tip. And there it stands. I have had inimal influence on the price.

But I tell you, my good friend, and you trust me, so you buy. You can only afford to risk $20,000 on this inside tip. We have had minimal influence on the price.

You tell dsquared and he trusts you, so he buys. He can only afford to risk $5,000. The three of us have had minimal influence. Dsquared tells rortybomb, who trusts him and buys. He can only afford to risk $500,000. Now we’re starting to influence the price a bit. Should it be considered public knowledge by now? Or is it still a secret?

As long as the bulk of the money is behind the public consensus view, a small minority who knows the truth will have small effect. I’m thinking the biggest effect will come when roughly half (weighted by #dollars) of the investors know the inside scoop and the other half doesn’t. From that point on the dwindling minority that doesn’t get it will have an increasingly smaller influence on the price.

So given a mass of people who have access to the public knowledge, a small minority that has correct special knowledge will not have a large effect. They are a contrarian small minority. They can presumably make money but they will not have a large effect on price.

Am I wrong?

Also, on a deeper level—the (few) arbitrage strategies I understand in any detail often seemed to proceed by assuming that small premiums people were earning for some particular type of asset were based on public but silly information, people assuming that asset was a little less risky than some other type that looked legally identical. Well, maybe the folk wisdom was right. Maybe people have some prejudice that looks irrational in normal times but is actually a small risk that will only show up in times of great financial stress. I remember someone claiming that LTCM hadn’t actually earned an arbitrage profit when it was going well, but had been fairly compensated for the risk involved in betting on obscure bond types issued by less trustworthy governments.

If you assume that it’s publicly-available information that you lack, why is it that everybody but you has access to that information and factors it into their choices?

What about liquidity risk? I never understood how it wasn’t possible that the arbitrage opportunities identified on the market weren’t real returns to tiny risks of a failure of liquidity in certain markets, or what instruments would be safer in case of a big financial blowup. Actually, I never understood how you could serenely say that a seeming mispricing wasn’t because of lack of information—you just didn’t know something subtle about the seemingly underpriced asset. I just don’t understand how real actions guided by EMH can rule this out.

If you get really interested, the papers listed here are a solid basis of the debate as it was 2 years ago:
http://www.sfu.ca/~kkasa/econ482.html

Dsquared: Great comments, but I’m still not convinced that autocorrelation and other related stock movements aren’t just proxies for risk (consistent with EMH). Look at the momentum chaser blowup in the hedge fund community in 2007!

Further, while there is room for technical improvements, the technical choices are pretty clear.

In electricity the fight is clearly between renewables, nuclear or a mix of the two. Fossil fuels with capture and storage is not going happen in the time frame and scale we need; with today’s technology or likely breakthroughs over the next few decades it will stay more expensive than both renewables and nuclear.

The means both of making new buildings more efficient, and of retrofitting existing buildings to improve their efficiency are also well known via insulation, sealing of ducts and of openings to the outside and so on. (Also more efficient lighting and appliances.) Once that is done, depending on various circumstances, ground source heat pumps or solar heating can supply much of remaining demand – further reducing building demand.

Similarly in transporation the answer is obvious – electric trains for most freight – with trucks use only to get goods to an from rail freightyards. Simlarly electrification is the answer for passenger transport electric cars, plus to some extent electric trains , the latter obviously where populatin distribution supports it. (Yes, bikes, walking, more telecommuting too – but mainly electric cars and trains.)

In the above sectors, emissions pricing will help, but mainly as a way of mopping up.

Because industry involves such a huge mix both of means and of ends, emissions pricing/permits will have to be the major means of control . Though even here there are some clear choices. About 70% of industrial consumption is in boilers – so various means to reduce emissions from these will be needed. (Or where this can’t be done, waste heat from such boilers can produce electricity as a side effect.) Much of the remaining energy not used in boilers drives motor, and means to make these more efficient are well know. A high percent of industrial motors are used to drive pumps, and again we know how to make pumps much more efficient. So clearly there is room for effciency regulations to improve industrial effciency. But it is also worth remembering that a lot the ways to improve industrial efficiency require more cleverness than simply improving key processes. For example, a lot of steel making is close to maximum thermodynamic efficiency. But you can still improve industrial processes to reduce scrapping during steel manufacture and during steel use. And for that matter you can substitute other stuff for steel, sometimes with huge savings in energy and improvements in quality. This is the kind of thing markets are good at and rule based regulation or subsidies do badly. That is why in industry, unlike other sectors, reductions will have to be reduced via emissions taxes or auctioned permits.

As someone who is not an economist but has a very strong math background, how hard is it going to be for me to gain anything from these articles?