Election Exercise Part II

Last Saturday I discussed the ‘49%’ rule for estimating the vote given current polls. The idea centers on a proposition that undecided voters tend to break for the challenger. I promised that I would do the analysis again today as well as Election Eve so that we can see how this strategy compares to the real totals.

I also decided to include (and I will include these same measure on Monday as well) other estimates based on the same LA Times numbers.

Method A: States where leads within the margin of error are called ties
Kerry: 153 Bush: 168 Tie: 217

Method B: All leads are counted as wins (despite margin of error)
Kerry: 285 Bush: 219 Tie: 34

Method C: Method B with reasonable guesses based on conventional wisdom/history for tied states (Bush: CO AR Kerry: HI, NJ)
Kerry: 304 Bush: 234 Tie: 0

Method D: 49% rule
Kerry: 330 Bush: 191 Tie: 17

Method D – last Saturday's 49% rule for comparison
Kerry: 310 Bush: 182 Tie: 46

Note: This is just for fun. Technically the only method that is statistically significant is Method A. Though not as interesting.

There is a lot of wrong information about polls that the media picks up on and helps distribute. It’s not because the media is evil or picking sides, it is because the truth of polls is that the results are usually not ‘sexy’. For example, if one candidate is leading the other candidate by 5 percentage points in a poll but the margin of error is +-4 points (most polls with 1000 people or less), then the correct interpretation is actually “we do not have enough evidence to suggest that the public has a preference of one candidate over another”. Boring, huh? Sounds better to say one candidate has a 5 point lead!

For consistency, below is today's electoral map applied with the 49% rule.

Pondering Social Security

There has been a lot of discussion of Social Security lately among the candidates. Not much of substance really but the issue has been raised. It has got me thinking. I know that the the conventional wisdom is that 1) social security doesn't provide much and 2) social security is going away as the baby boom generation starts to enter retirement.

I decided to try to understand the dynamics of social security a bit better for myself and crunch some numbers.

Basically I found that with an inflation rate of 4% and today's mix of worker/retirees there is an 'effective' rate of return for an individual of 4.64%. (My assumptions and model can be found in an appendix page). Now keep in mind, social security is not an investment plan - it's a distribution of money from the labor force to the retiree community - but the 4.64% interest rate represents an 'effective' interest rate that an individual would see over the years. This is interesting. Based on a lot of the rhetoric I kind of expected that this number would be a negative one (i.e. overall you lose money in the social security program). Actually it keeps up with inflation and, in fact, over performs it. Not bad. No great windfall for sure, but not bad. It does provide better than a bank account for those not wanting to understand personal finance. For us investers familiar with portfolio allocation - you can consider it a 'cash' asset within the context of your greater portfolio given it's volatility and rate of return dynamics.

Interesting observations to note:

1. the average salary as it changes over time actually has nothing to do with the calculated effective interest rate - it's just relative
2. the social security rate (7.65%) also has nothing to do with the calculated effective interest rate - it's also relative. What the SS rate has to with is the relative sizes of both the payment and the benefit but the overall rate of return remains the same.
3. the ratio of retired/working will ebb and flow over time - but responsible practitioners can adjust by either:

• truly holding a separate social security account that does balance in the macro - for example, every 30 years thus the changes in the ratio can be ridden out with surplus at times and deficit at times but evens out in the long run - or-
• adjusting some of the parameters on an occasional basis

The first option is certainly more politically feasible. Obviously if there are major trends in one direction in the demography, the policy variables should probably be adjusted (for example, if we are living longer, the retirement age should be upped).

I still like this program if it's administered correctly based on these results. I would also support a 'forced' 401k type program if that were on the table - which I guess is the notion of privatizing parts of social security.

My main concern is having a minimum amount of subsidence for those that made poor financial decisions in their life (I actually have no confidence in the general population making responsible financial decisions based on my observations) so I won't be burdened by them in my retirement (which I have not only spent the time to learn personal financial planning but also more importantly am committed to delaying instant gratification to responsibly fund my retirement).

Interestingly, an untalked about benefit of this system is that the longer you live, the higher your 'effective' interest rate is because you keep getting benefits until you die. This is exactly opposite any direct investment plan. The longer you live in a direct investment plan, you will need to get a higher rate of return from your investments to fund your retirement properly. Since you usually can't plan on your death, it's kind of nice to have both vehicles and dynamics working for you.

Conclusion: The Social Security system is not a great evil. It's certainly useful if it's administered correctly by our politicians. We need a commitment to holding it in one account (pretty simple, huh?) as well as occasionally adjusting some of the parameters as demographics change. Private investment accounts aided by the government would certainly have it's place as well and I would welcome any more help (thank you for the 401k!!).

Out Of Step Time Zones

I think that a time zone choice says a lot about a country and its leadership. Take India for example. India has chosen to have only one time zone for the whole country. If you look at a map, they really should have two. Country leaders, however, decided that they only wanted to have one time zone for the whole country so they split the difference and created one large time zone that is a half-hour off from the other time zones of the world. India leaders made the decision to spit in the face of world standards so that their citizens wouldn’t have the luxury that we all enjoy of adjusting our watches when we fly domestically.

I think this tells us something about them.

If you think that India’s decision is a bit off, consider that there are a few more oddballs. Burma, Iran, Afghanistan, Nepal, Central Australia, and Newfoundland (part of Canada) are all off by a half hour as well. These countries / zones are not near as wide as India so they don't really have any excuses. It gets stranger. Nepal and the Chatham Islands of New Zealand are both off by 45 minutes. There are also some small island groups in the Pacific that have their own versions of local time (which actually nobody really cares anyways - do what you want).

Now I’ll be the first to admit that with the internet and sites like timeanddate.com it’s not too hard to figure out local times for any country or city anymore. But still. The reason that the time zones were adopted in 1884 was to help facilitate commerce and coordinate train schedules throughout the world in an efficient and fairly simple way that benefits everyone.

I say to all of you countries and regions that think you are more important than the world community – what kind of image of yourself are you sending? Come join us and become better world citizens.

Movie Review - "Sideways"

I subscribe to a movie screening series called Talk Cinema. Roughly every three weeks on Sunday mornings we get to screen a movie that has not been released to the general public yet. The rub is that you don’t know what the movie is going to be until you show up. In addition, at the end of the movie they have a guest speaker (usually a local reviewer but we have had the director on occasion) who leads a discussion of the movie. These movies are usually more of the 'independent/foreign' movie variety and I have never been disappointed by their selections.

So far in this fall season we have had three films. The first was “I Love Huckabees” and the next was “Primer” (by a first-time Dallas filmmaker). This morning the movie was “Sideways” starring Paul Giamatti and Thomas Hayden Church as two early middle-age college buddies spending a week together before the Church character is scheduled to get married.

The movie is surprisingly deep and just as surprising – very funny. It takes you through a world of wine tasting and Santa Barbara wineries that make you want to run out and try the latest Pinot Noir. Paul Giamatti is as always tremendous (you have seen him playing Harvey Pekar in American Spendor) playing the neurotic ‘loser’ with his usual complete believability. Church (you probably know him as Lowell on the TV series “Wings”) is just as convincing as a never-grow-up ladies man.

This movie is ultimately about hope – getting over hurt and back into life. The wine metaphor is everywhere – even the title is taken from how a bottle of wine should be stored for best preservation.

I give a thumbs up – way up!

Election Exercise Part I

There is a paper written that suggests that raw polls are not very accurate if you only look at the comparative results. But, this paper goes on to say, in elections where there is an incumbent you can analyze the polls in such a way that historically has been fairly predictive. Basically they say to simply look at the incumbent's percentage of the poll. If the number is below 49% then they predict victory for the challenger. If the number is above 49% they predict victory for the incumbent. If he is at 49%, they say it is too close to call. Central to the theory is that a large percentage of undecided's historically go to the challenger.

So I will test this theory by using the LA Times Electoral Vote Tracker to do an analysis. For every state that they have 'tied' I will use the 49% rule to make the decision. The obvious shortcoming to this analysis is that many of these states polls are not statistically significant enough to make this distinction. But - what the heck. Let's see how it does anyways. We're doing this for SCIENCE!! I will do this today, this coming Friday, and finally once more on Monday right before election day.

Today's results would be the following:

Kerry: 310 Bush: 182 Tie: 46



To be continued...

Ode To October

I love October. I've told friends before that in all my travels, October seems to be the most consistent around the world for good reliable weather. An added benefit is that travel is low-season since it falls between summer break and the holidays. So there are usually great values to be found. If you are going to travel, October is a prime month for it. In Dallas, I must admit I wish we had more fall colors but besides for that the weather is good! To complete the 'perfect storm', October also ushers in football season, hockey season (well, not this year), opera season and theater season. Some folks say that springtime is the time of renewal and fresh starts. I submit to you that it is October.

Runners up include June and December. June ranks up there because of the plethora of fruits and berries available. I call it my fresh eating month. Also it is the month when my crepe myrtles show their stuff in red brilliance. Finally, it's hard to discount December because it is rich with holiday cheer and business starts to ratchet down a bit.

Friday Spinach

This morning (and from now on - EVERY Friday morning) the office has Krispy Kreme donuts for us. It makes it very hard to walk into the break room to get my fruit juice and vegetables - which I somehow rarely do when other goodies like candy and donuts and hydrogentated oil-laden microwave popcorn is available.

I hate to look a gift horse in the mouth but I am not happy about this Krispy Kreme development at all. Instead I vote for free Friday spinach or maybe free Friday canned mackerel. Just imagine how we could eat and socialize knowing happily that we are also getting good nutrition. ;-)

Death Of The Bambino...?

.....well maybe not yet but can you believe that that the Red Sox won 4 straight to beat the Yankees 4 games to 3? Apparently this series made major league baseball playoff history as no other team had come from behind like that.

I can't help but wonder how much productivity took place in Boston today...

I just found out that St. Louis beat Houston in the National League... Whew... that would have been very weird - Bush's home turf (Houston) vs. Kerry's home turf (Boston) in the World Series.

Go Red Sox! It would be nice to see the team get a world series after what, 86 years?

The Battle Of Campaign Websites

Continuing on the political theme, I took a moment to scan the two presidential candidates' websites. While there are many similarities, there are also some striking differences. For example, one of the first things you see when you go to the George W. Bush site is information about John Kerry and why he isn't fit to be president. In fact, on every issue statement on the Bush site, they contrast the position to Kerry (obviously with spin!) in a very easy to see format. The Kerry site, on the other hand hardly mentions Bush. They also have a bit more detailed information about forward plans and policy statements but don't provide direct access to the contrast to Bush.

I actually would have thought that these would have been the other way around. As the challenger, it would have seemed that Kerry would have more of a direct approach to Bush's failings on each of the issue statements.

I have to give the winner of this battle to the Bush site. Visit both by hitting the graphics above.

The Mobile Home Vote

I have done a bit of analysis on the current electoral position of the 2004 Presidential Race. The data clearly shows that while the nation is very much divided between our candidates the following holds true.

Kerry is favored by those states that have:

• higher personal income per capita
• higher percentage of college graduates
• higher percentage of professionals and managers
• better quality primary and secondary education
• lower percentage of teen births
• lower percentage of mobile homes
• lower percentage of obesity

Bush correspondingly is favored by those states that have:

• lower personal income per capita
• lower percentage of college graduates
• lower percentage of professionals and managers
• lower quality primary and secondary education
• higher percentage of teen births
• higher percentage of mobile homes
• higher percentage of obesity

Conclusions: I just thought I saw some empirical trends and figured I'd take a stab at analyzing a bit further. Causation / Correlation / Relevance / Spin? I'll leave it the reader. :-)

Details in Appendix.

First Post

Well, I suppose that there must be a first post so here it is. Welcome to Random Thoughts where at random intervals I will share my thoughts. Thanks to Blogger.com for the software that makes this easy.