Twitter roundup

Comments

[David in Cal]

Are we really arguing over whether "no chance" literally means 0.00%?

I'm right with Silver. In fact, part of my actuarial study involved reading Knight, the author he recommends on Risk and Uncertainty.

One way to look at the difference between risk (measurable) and uncertainty (not measurable) is that "risk" is the chance of a bad result, assuming a certain model is valid. "Uncertainty" is the chance that the model isn't valid for the current situation.

Alternatively "uncertainty" is related to the unk-unks (the unknown unknowns.)

Here's another way to understand risk vs. uncertainty. Here are five possible things Brendan might mean, when says Cain has "no chance" to win the nomination:

1. Casual way of saying he considers it unlikely. Like a schoolboy saying the Cubs have no chance to win the pennant next year.

2. Brendan has a political science model that says Cain's chance is very small, say less than 0.5%.

3. The model in #2 is accepted for publication in a reputable journal.

4. Brendan blogs that Cain has "no chance", risking embarassment if he turns out to be wrong.

5. At Intrade, Cain's odds are shown as 5.2%. Brendan is so certain that he bets half his savings that Cain will lose. He figures it's "free money" or what Lloyds of London informally calls "money for old rope".

P.S. Silver quotes a speechwriter's promise that he would “quit political opinion writing forever” if Mr. Cain is nominated. That reminds me of an old joke about a marriage proposal.

The young lady asks in amazement, "If I turn you down, will you really commit suicide?"

The suitor sternly replies, "That has been my usual practice."

[Rob]

I join in Bob Somerby's strong recommendation that you watch the video of former Mo Dowd "research" assistant and current New York Times political blogger Ashley Parker. She's like, a total, like, airhead. If there were an ounce of justice in the world, Penn would refund the money she paid them for her, like, education.

[David in Cal]

I might add that I've had personal experience with the situation described above. I once wrote a paper about something called Table L. Table L shows the exact risk for certain workers compensation insurance policies. The paper easily met the formal standards for being valid: it got published in a referred journal, it won an award, and it has been on the Casualty Actuaries' syllabus of required reading for 35 years.

However, from a real-world POV, Table L is not accurate. It ignores some factors that I know are significant and it may be affected by unk-unks as well. I would not actually sell insurance policies based on Table L.

[David in Cal]

I am disgusted by the level of irresponsibility, when Perry asserts, based on no evidence, that his tax reductions will pay for themselves. Ditto for Cain's tax plan.

A similar level of irresponsiblity is manifested by our President. He keeps imposing spending increasing, including some (like student loan forgiveness) which are not authorized by Congress and thus apparently illegal. He claims that higher spending will somehow reduce the deficit. Ditto the Dems on the Supercomittee, whose deficit-reduction plan contains considerable spending increases.

Where are the grownups?

[Metrichead]

Don't forget David, the student loan forgiveness plan really isn't "forgiveness." The debt remaining after 20/25 years that's "forgiven" is actually considered "income" and gets taxed by the IRS.

[David in Cal]

I hadn't thought about that aspect, Metrichead. Let me try a hypothetical example to be sure I understand.

Suppose someone's loan balance of $50,000 is "forgiven". That adds $50,000 to his/her taxable income. So, if s/he's in a 20% tax bracket, she'll owe an extra $10,000 in federal income tax. So, the net benefit to the individual (which equals the net cost to the federal government) would be $40,000, rather than $50,000.

[David in Cal]

The main problem with WaPo's article about Rubio's Vice Presidential value isn't that VPs seldom make a difference. It's that WaPo is doubling down on a smear. It repeats a false accusation that Rubio embellished his background. The Miami Herald took WaPo severely to task. AFAIK other major news sources have been professional enough not to support WaPo's smear. However, WaPo is evidently commited to it.

BTW contrary to this WaPo editorial, it's widely conceded that Rubio would be an ideal VP candidate, particularly for Romney: Rubio is bright, conservative, good speaker, good looking, and, of course, Hispanic.

Another problem with WaPo's article is that it omits the fact that Rubio has declared that he's not going to run for VP:

Speaking at a political forum in Washington, Rubio said flatly on Wednesday: "I'm not going to be the vice presidential nominee." He then repeated himself.

Rubio said he didn't run for the Senate "to have a launching pad for another job." http://www.chron.com/news/article/Rubio-says-he-won-t-be-VP-nominee-2204728.php

[Metrichead]

David - the short answer is yes. I have an old friend who I went to college with has 8 years of school in total = 5 years undergrad from Eastern Michigan and 3 years of law school from Penn State, and has accumulated $275,000 in debts. 3/4 of that $275,000 are from federal Stafford loans.

So about $206,000 can be forgiven after 25 years, but the remaining balance forgiven is considered income and will be taxed in addition to whatever you're currently making.

If I'm not mistaken, you cannot file for bankruptcy for student loan debt. But if you go to work for public service or a qualified non-profit, the debt can be forgiven in 10 years with no tax liability.

[David in Cal]

Here's another question, Metrichead. Suppose someone, like your friend, knows his loan will be forgiven in 25 years. Could he just make the minimum payments and wait for the forgiveness? Since he knows the loan will be forgiven at a certain point in time, it would seem that he doesn't need to try to pay the entire loan off.

If this is a realistic picture, and if a lot of people use this strategy, it seems that the cost to the government could be quite high. However, I'm not familiar with the structure of these loans, so I don't know how realistic my hypothetical situation is. Do you know?