Monday, November 5, 2012

Election night clues

I've been following the presidential polls more closely recently and here's my take on what to look for on election night.

As of 3 November, CNN has President Obama with a lead in the electoral college, and classifies the following eastern states as "toss-ups": New Hampshire, Virginia, Florida, and Ohio. Almost all the analysts believe that Governor Romney must take at least two of these states to have any chance of winning. They also point out that no Republican has ever won the election without carrying Ohio.

Note that although I'm using the CNN map below, others (Rasmussen, Real Clear Politics) pretty much agree with this assessment. Here's the map:

Map courtesy CNN



Looking only at the toss-up states (I'm making a large assumption that each candidate will hold onto the states that are either safe for them or leaning towards them), currently:
  • All of the polls in New Hampshire show President Obama with a small (2-5 point) but stable lead.
  • All of the polls in Virginia show President Obama with a 3-5 point lead.
  • In Florida, things are more confused. Three polls show President Obama with a 2-point lead, while one poll shows Governor Romney with a 6-point lead.
  • All the polls show President Obama with a lead in Ohio.
So, when the returns start to come in, watch these four eastern states. They'll be your clue as to how the night will go.
  • If Governor Romney takes all four states, that gives him 270 electoral votes, and he wins. That's pretty far-fetched, given the polls I mentioned earlier.
  • Governor Romney needs to win three of these four toss-up states to have any chance. If Governor Romney takes Florida and Virgina, he still has a small chance. But he'll also need to win almost all of the other tossup states. If he also takes Ohio, that will help him a lot.
  • If Governor Romney loses three of the four states, the election is over.

Wednesday morning update


Obama won the election and took New Hampshire, Virginia, and Ohio. Things are too close to call in Florida, but Obama leads there as well.

If Obama holds his lead in Florida (most of the uncounted ballots are from in Dade County, which leans heavily towards Obama, so it looks like he'll win that state), Obama took all of the toss-up states. Impressive.

Saturday, November 3, 2012

Obstructionism and the Filibuster



During the Bush administration, Democrats were frequently accused of being "obstructionist".

Fox News published are article just before the start of Bush's second term headlined "Dems Plan to Obstruct Bush Agenda." Six month later, President Bush accused them of obstructionism himself:

"President Bush spent Tuesday ... accusing Democrats of standing for nothing but obstructionism."

All this despite the fact that until January 2007, the Republicans had firm control over the White House and both Houses of Congress.


One of the tools that the minority in the Senate can use to block the will of the majority is through the use of the filibuster. In the "olden days", that was romanticized as a lone Senator standing on the floor and arguing his position, and refusing to yield the floor until he or she either wore and and gave up , or until the majority gave in. These days, it's much easier; the minority just signals that they're going to filibuster the bill and don't actually have to say anything.

Overcoming a filibuster requires a vote of cloture. In order to pass the vote and end the filibuster, a three-fifths vote of the Senate is required. In the modern Senate, that's 60 votes.

The filibuster - not so romantic any more


There are no direct counts of the number of filibusters conducted. The Senate does, however, publish the number of cloture votes held, which under-represents the actual number of filibusters, because there are some filibusters that don't go to a cloture vote because the majority concedes that they don't have enough votes to overcome it.

So we're forced to use the number of cloture votes instead of actual filibusters.

Here are the number of cloture votes by Congressional session, with the minority party noted. As can be clearly seen, the number of filibuster roughly doubled when the GOP became the minority party in 2007.





Source: US Senate



Clearly, the GOP has been more obstructionist than the Democrats were. The number of cloture votes in the last six years since the Democrats took majority control of the Senate has doubled.


Friday, November 2, 2012

The Labor Participation Rate and unemployment

I was talking with someone a few days ago about the unemployment rate, and pointing out that in his first term, President Obama has a better record on unemployment than both Presidents Reagan (see it here) and George W Bush (see it here) in their first terms. This person claimed that the unemployment rate looks artificially better than it really is because of a drop in the Labor Participation Rate.

Defining terms


The unemployment rate is the percentage of the work force that's out of work, but looking for work.

The Labor Participation Rate (LPR) is the percentage of the general population of the country that's in the work force.  You aren't in the work force if (1) you're under 16, (2) you're institutionalized (in a nursing home, in prison, or in the military), (3) you aren't available for work (retired, a stay-at-home parent, disabled, etc.), or (4) you haven't looked for work in more than a year.

The LPR is falling! The LPR is falling!


Back to the story. The person I was talking to claimed that the LPR is dropping because so many workers have grown discouraged and have left the workforce. He pointed out that the LPR has dropped from 65.8% when Obama took office to 63.6% now. And, he said, lowering the size of the work force makes unemployment look lower than it really is.

So let's look at those two claims: (1) the reason people are leaving the workforce is that they're growing discouraged, and (2) a smaller work force makes the employment situation look better than it is.

Why is the LPR dropping?


Data on the LPR is available from the Bureau of Labor Statistics (BLS). The data does indeed show that the LPR is dropping. However, it shows that the LPR has been dropping since 2000.

Here, let's look at a chart of the LPR from 1960 to the present:

Labor Participation Rate 1960 - present (chart courtesy BLS)





















Does that shape look familiar? It should, it's the first part of the classic bell-shaped curve. It shows that the LPR began to increase about 1965, reached its peak in the late 1990's, and began to decline in 2001.

Are there anything that can account for that bulge? Of course, and it should be obvious - the Baby Boomers. Boomers were born between 1946 and 1964. The bell-shaped curve begins it upward sweep when the first Boomers joined the job market in the mid 60's, and it began declining when the first boomers began retiring around the turn of the century. Now, about 10,000 Boomers are retiring every day. From the link:
A recent article by the Chicago Federal Reserve Bank estimated that at least half of the decline in the labor force participation rate (LFPR) in the recent recession was due to retirement.
I'm not saying that none of the LPR drop is caused by the recession, but the bell-shaped curve caused by the Boomers leaving the job market is clearly a major factor.

So I think I've established that the drop in the LPR isn't caused by the recession (although the recession does have some impact).

Now, let's turn to the second claim. Namely, that those who retired in discouragement aren't counted as part of the work force and are therefore making the unemployment situation looks better than it really is.

I'm discouraged. Do I count?


Remember the definition of terms up above? If not, scroll back up and take a look at who's defined as not in the work force. It's OK - I'll wait.

Welcome back - let's continue. The 4th reason for someone not being in the work force is that they haven't looked for a job in more than a year. As long as they want to work, are available for work, and have looked for a job at any point any time in the last year, they're counted as part of the work force. These folks are called "discouraged workers" by the BLS.

As I've said before, the BLS reports on several unemployment statistics every month. Among them is the report on U-6 unemployment. U-6 includes those discouraged workers. And it, like the more frequently reported U-3 ("official") unemployment report, has been dropping. In the last year, it's dropped a full percentage report, from 15.7% to 14.7%.

So the only people who would cause the LPR to go down and who wouldn't appear as unemployed are those who meet all these conditions:
  • Over 16
  • Not in a nursing home, prison, or the military
  • Unemployed
  • Want to work
  • Are available for work
  • Haven't applied for any job in more than a year
I know a few unemployed people, but I don't know anyone who meets all those criteria. I suspect they're pretty rare. If anyone out there can find any hard data one way or another on this subject, I'd love to see it.