ECON 439: Public Policy Analysis

Last week during President Obama’s State of The Union address he made the claim that women workers make 77 cents for every dollar that men make. It was one of the President’s largest applause lines indicating most in the audience saw the wage discrepancy and as area for change. Politicians and media members tout this statistic over and over as evidence of discrimination against women. With all the progress this country has made in the last 50 years it’s hard to believe society would tolerate such unfairness. We’re living in the 21^st century after all, get your stuff together U.S.!

Where did the President come up with his 77 cents figure? Basically from looking at U.S. Census data which says “the average man in the U.S makes X and the average woman makes Y.” But census data doesn’t control for differences in work experience or education. Probably most importantly it doesn’t control for the types of jobs that men and women work. One of the largest contributors to the difference in pay is that men and women make different decisions when it comes to the types of jobs they’re willing to work. Men are more willing to work on high voltage power lines than women are. Men are more willing to work on crab boats too. These jobs are dangerous and pay significantly more than would a job as a teacher or administrative assistant. I read a study recently that controlled for these various factors and found the wage difference to be between 9 and 5 cents per dollar and there were still variables not controlled for that may have been able to explain this small difference. In fact, when these other variables are accounted for women often make more than men. This makes sense to me intuitively. If I assume that business owners are rational and that profit maximization is top priority, what incentive would one have to hire men at all? If one could hire a woman to do the same job a man does for 25% less it would seem women would outnumber men in the workforce as this would create more producer surplus.

On a day as forced and manipulated by corporations and multimedia companies as Valentines, there are deals galore from chocolate, to booze. But does this drastic shift in prices correlate with the season of those in love, or the loveless? Many of my friends love to celebrate their alone-ness on a day that memes usually depict single folk as miserable and drowning in ice cream not beer. But due to recent trends, especially in area as young and rambunctious as Corvallis, the trend is to celebrate single status rather that begin self loathing as soon as the first pink side-stacks filled with heart shaped things begin appearing. But in reality everyone has enjoyed the goodies since their elementary school days, and now alcohol is a substitute for the sugar high we all use to get as kids. With a market demand seemingly revived by the first major corporate holiday since the Christmas and new years season, companies have obviously taken note of the demand. even if the season is directly correlated with the idea of true love, relationships, and spoiling your significant other; beer and other booze is still bought by those celebrating themselves and their friends. You can say the corporate love machine accidentally created a separate holiday by accident. Cheers.

Tomorrow evening, millions of people will tune in to watch the forty-eighth Super Bowl between the Seahawks and the Broncos. Some will do it because they are fans of one of the teams that are playing, many will do it to watch the commercials. But a few will tune in, or at least care about the outcome of the game for a very different reason. The stock market. In 1978, sports columnist Leonard Koppett jokingly published a piece claiming that whether a NFC or an AFC team won could predict a bullish or bearish respectively over the following year. Despite the intended satirical nature of Koppett’s column, the Super Bowl Indicator or SBI has correctly predicted the market 80% of the time since it was first published. So is this a legitimate predictor of the market? No. A rudimentary understanding of the principle that correlation does not imply causation would cause a person to dismiss this indicator swiftly.  A possible explanation for the correlation of these events is the observer effect. The indicator becomes a self-fulfilling prophecy as people take the score seriously and react in accordance with it.

This article, from a British national newspaper, caught my eye yesterday:

http://www.telegraph.co.uk/news/uknews/crime/10607071/Fifth-of-unemployment-benefit-claimants-have-criminal-record-say-new-figures.html

The article describes a study done by the UK government which compared information contained in court records and national law enforcement databases with names appearing on government assistance rolls. The results found that about 22% of people claiming unemployment-related benefits have had some kind of criminal conviction or have served prison time previous twelve years.  The first thing I thought when I read the headline was that there’s no way to even begin to sort out causality with so little information.  Do unemployed people commit more crime or are people who commit crimes more often unemployed? Or is there something that unemployed people and criminals have in common that is unobservable in this study? I also found it interesting that the headline (if not the original study itself, which I haven’t yet located) only mentions a portion of welfare recipients.  Why is that? A higher percentage certainly gets you a jazzier, more shocking headline. Perhaps if recipients of other types of welfare benefits were included, (things like benefits for low-income parents, or disability payments) the numbers would be much lower. In fact, the article mentions that when the search was expanded to include other types of benefits – such as the state pension, the figure dropped to 7%. That was not surprising to me at all. Generally speaking, there’s a sort of “aging out” process in crime. Younger people commit the larger proportion of crimes, and once people who do commit crimes get older, they tend to cease or at least greatly reduce their criminal activity.

Another thing that that occurred to me while reading this was the different ways that information like this can be used to influence both public opinion and policy makers. The Telegraph is known for being a very conservative publication – even so, I thought that the tone of the above article was relatively even-handed. Depending on your goals, however, you could either use a study like this to demonize some groups of people and justify cuts to their assistance, or to highlight the problems that these same groups face and use it to justify increasing assistance. Overall, a study like this presents more questions that it really answers. Crime, poverty, and the State’s responses to them are so complex that they can’t be boiled down to a single headline. (Of course, that doesn’t stop media outlets from trying!)

– Brandy

For the travelers out there,

It is more than likely you have witnessed how locals often despise tourists. I wonder if this is a correlation or causation scenario. As I sit here before my pint of beer pondering this question, I reflect on my time traveling in Europe this summer.

Many countries I visited didn’t like tourists, especially in countries with more tourists floating around. This was a common scene:

There were plenty of people with their iPads, taking pictures in front of inanimate objects that they may not fully understand the significance of. When I spoke to locals this often irritated them beyond belief.

The argument for correlation:

Maybe, just maybe, the people of these countries are just bitter people. The locals of countries seem to have been less tourist friendly than countries farther north. Perhaps their cultures are more reserved in regards to showing positive emotions [I am doubting this]. I mostly write this as a means of discrediting that it is merely correlational.

The argument for causation:

My theory is this is due to the locals being forced to work and ignore their wonderful countries while other people come in and have all the fun. They see people running around not speaking their language and acting pompous. Of course this would frustrate the masses. Thus, these places with beautiful climates and more sites (such as Italy) attract higher amounts of tourists. The residents of countries with less tourists likely are more receptive to tourists as they are not constantly swarmed by them.

Verdict: This argument is an eternal one, but my vote is for the causation. Share your thoughts and remember to upvote this post.

Most people will associate 1.6 billion people with the word “China.” Chinese Spring Festival, the most important holiday in the year, can make over 250 million people to use public transportation to go back to their hometown. Those people who work away from home would like to choose train as their vehicles. And they need to wait in line and spend an average of five hours to get a ticket. Therefore a few people see money for those train tickets. These people are called “ticket scalper”. Ticket scalpers usually hoard a mass of the train tickets so that most citizens couldn’t get the tickets. Thus the ticket scalpers could sell those tickets with higher prices. But this conforms to the rules of market economy. A large number of consumers who would rather spend three or four times higher than normal prices on the train ticket but not wasting five or six hours waiting in line result in the appearance of ticket scalpers. Those ticket scalpers’ primary targets are the migrant workers who are going to return to their hometowns. It makes lots of workers couldn’t go back to their hometowns, because they couldn’t afford the higher prices. In order to stop this “fail market,” the government decides to apply for the real name and ID system. Each person can only buy a ticket with his or her own ID card. This government intervention results in the decrease number of ticket scalpers, which could lead to the increase of the unemployment rate and the increase of the risks of social instability in other way, but it ensures a great number of migrant workers to go home and get together with their families in the spring festival with fair normal prices.

http://www.mayoclinic.org/healthy-living/nutrition-and-healthy-eating/expert-blog/red-meat/bgp-20056277

There seems to be more and more literature on how bad red meat is for humans and the increased risk of heart disease it causes.  Does red meat really cause heart disease or is it something else?  My paternal grandparents both grew up on farms in Nebraska.  My grandfather died in 1981, at the age of 81, of a brain hemorrhage due to a defect in an artery, and my grandmother died in 2008, at the age of 104, from old age.  Both had grown up eating red meat almost every day.  When I spent my summer vacations with them, we would eat a lot of red meat.  My grandparents raised beef and chickens so there was almost always eggs for breakfast and some type of red meat for dinner.  Neither of my grandparents had any issues with heart disease and as far as I know, their siblings didn’t either.  I think there may be something else in the meat we buy at the store that is causing heart disease.  Could it be the type of diet the cattle are on?  Are the cattle farms near large cities which have a lot of smog?  Is the water contaminated?    I think there are too many variables to specifically state that red meat causes heart disease.  I think that a good study would be to see how health compares in people who eat the same type of diet except some eat red meat from environmentally clean farms and the others who eat red meat from stores.  There is definitely a correlation with red meat and heart disease but I’m suspect to say that red meat causes heart disease.

http://www.dailymail.co.uk/sciencetech/article-2069018/Teenage-sex-affects-mood-brain-development-reproductive-tissues-later-life.html

This article I found tries to show a link between early sexual activity and bad mood later on in life. To study and prove this correlation the researchers used hamsters as a replacement of humans. The problem with this experiment is that hamsters and humans have vastly different brain chemistry and humans are much more emotionally intelligent. Now besides this huge issue, we can also see that the correlation found (that when the younger hamsters had sex vs. the older, the young were more likely to act “depressed” or “anxious later on in life.”) does not imply causation.

While the researchers did observe a change in the younger hamsters behavior, there was no information on whether or not a control group was used to see if young hamsters without sexual activity were also anxious. Another factor playing into these hamsters moods and actions is simply age. The experiment followed a group of 40 day old hamsters and a group of 80 day old hamsters, therefore the 80 day old hamsters were always older. What should be considered is maybe younger hamsters are generally just more anxious than older hamsters, and activities such as swimming or running through a maze make younger hamsters act differently than any older hamster would, regardless of previous sexual activity.

Making this link with the hamsters and then also implying that “teenage sex leads to bad moods later in life” for humans is not only presumptuous but it is also bad experimenting. If a study wanted to prove this they should attempt to study humans or use an animal with a higher intelligence.

I am from Washington and I am one of the many people that live there that has quite bad allergies. Without pseudoephedrine, an allergy medicine, my allergies are out of control at certain points in the year. When I was younger it was not a problem because my mom could go to the pharmacy and pick up as many boxes as needed. However, people began abusing that privilege and started buying the drug to make meth. In order to stop this abuse the state of Washington decided to move pseudoephedrine to behind the counter. Now you can only purchase a certain amount of the medicine and must show a valid ID at checkout. This has significantly hurt sales at pharmacies but it also has decreased the number of meth labs in the state. Each state has a different policy on the purchase of pseudoephedrine. For instance, here in Oregon people cannot get pseudoephedrine without a prescription from their doctor. This law went into effect in 2004 and since then the amount of meth labs has decreased by 97%. This policy has a very clear outcome. So prior to 2004 people were able to buy as much pseudoephedrine as they felt necessary, which then caused people to abuse the drug and use it in meth. This then caused the government to put a restriction on the drug, which resulted in a large decrease in meth labs. The causality that resulted from this policy is very clear and hard to argue.

People often get confused with correlation and causation. Because variable A is correlated with a variable B, that does not mean A causes B. It could mean that B causes A, A causes B, or it could be a case of omitted variable bias. For example: a high college GPA (dependent variable [y]) could be correlated with a higher high school GPA (independent variable [x]). When doing a regression analysis a person can perform a variety of tests to prove if X is correlated with Y. The most basic step is to find the value of R^2, which measures the fraction proportion of the sample variation in Y explained by X. If a researcher gets a high R^2, this is good but still does not mean there is causation. You can estimate how much causation this is by finding the marginal effect X has on Y. For example, if you get a value of .08 for variable X this implies that an increase in high school GPA is associated with an increase in college GPA of 8%. The next step would be to run an OLS (ordinary least squares) estimator on the population regression function. The purpose of the OLS is to find any bias in the variables you are comparing. This also takes into consideration any unmeasurable variables (error terms) like motivation or ability.

According to act.org, first year college GPA and high school GPA are directly associated with each other. It has a much larger direct effect on first-year college GPA than it has on degree completion within 6 years. This analysis did not take into effect any unobserved variables which could have an effect on the results.

Reference:

http://www.act.org/research/researchers/briefs/pdf/2013-8.pdf