Tuesday, May 17, 2022

Learning What You Pay For?

The chart below shows the top 30 colleges according to US News rankings and their respective tuition and fees. Schools below the linear regression are theoretically a better than average deal relative to the group, while those above the line might be supposed to be a worse than average deal. (This assumes a uniform distribution in quality, e.g. the difference between the first and second rank is the same as the difference between the second and third, etc.) Five of the top ranked schools are public: University of Michigan - Anne Arbor, University of Virginia, and the Los Angeles, Berkeley, and Santa Barbara campuses of the University of California. For these schools the out-of-state tuition rates were used. In some cases two or more colleges tied for a rank. It would appear that Princeton is the best deal at $56,010, while Tufts is the most overpriced of the group at $63,000. 





 

Wednesday, March 9, 2022

Crime in Minneapolis and Crime in the Nation

Has crime in Minneapolis gotten out of control? One side of this debate cites crime statistics that are far greater than (recent) historical averages, while the other side argues that crime has increased throughout the country and that attention to local policy is therefore unfounded. The chart below shows the change in homicide rates from 2010 to 2020 using 2010 as the baseline. Overall, the national rate increased by 35% over this period, while the Minneapolis rate increased by 102%. The Minneapolis chart shows more volatility, but the change from 2019 to 2020 still appears to be a divergence. Comparing Minneapolis to other cities may also be useful.


Figures for United States Data:

0.00%
-2.08%
-2.08%
-6.25%
-8.33%
2.08%
12.50%
10.42%
4.17%
6.25%
35.42%
Figures for Minneapolis Data:
0.00%
-6.70%
10.78%
5.74%
-17.03%
27.83%
-8.90%
12.18%
-20.37%
13.86%
102.01%

Tuesday, November 9, 2021

Multiples and Returns of Nine Communication Services Stocks

I have previously compared multiples and returns for stocks without isolating for a specific sector. The results were not particularly useful. PEG ratio was the only multiple that predicted stock performance in the "proper" direction, and it didn't do so with much consistency. I thought that if I analyzed a single sector, the results would be more meaningful. I compared the same multiples and tracked returns from the opening price on June 28, 2021 to the closing price on November 5, 2021 for nine large/mega cap stocks in the communication services sector (dividends were added to the returns). The results were not much better. The linear regression for PEG ratio and price/forward earnings for 2021 were the only multiples that produced the expected negative slopes. PEG ratio predicted positive returns more consistently than price/forward earnings. It might be worth noting that six of the nine stocks had negative returns, but those data points should still be valid; a useful multiple should predict the stocks that will lose the least value during a downturn as well as predicting those that will gain the most in a bull market. 








Friday, October 15, 2021

Addicted to Low Interest Rates?

 The charts below show the product of the effective federal funds rate and the unemployment rate on a monthly basis. The Federal Reserve is tasked with promoting "maximum employment and stable prices". Generally, it will try to move the federal funds rate lower when unemployment is high, and higher when unemployment is low, but these charts appear to support my feeling that near-zero rates have become the default. I made three charts: one beginning with 1990, one beginning with 2008, and one beginning with 2009. The most recent observation had the effective federal funds rate at .08 while the unemployment rate was 4.8 for a product of .384. While the federal funds rate has gone lower than this a few times, never has it done so with such a low unemployment rate. 










Sunday, September 19, 2021

Will Choi Reduce Racial Disparity?

 On September 8 Ramsey County Attorney John Choi “announced that his office would no longer prosecute most felony cases arising from low-level traffic stops, an effort aimed at reducing racial disparities in the criminal justice system,” the Star Tribune reported. The article states that “his office will decline to prosecute felony cases — such as those involving illegal possession of drugs or firearms — if evidence is discovered during a traffic stop for a nonpublic safety violation like expired tabs, objects dangling from mirrors or a burned-out taillight. The policy makes exceptions for cases that endanger public safety.” I'm struggling a little with Choi's apparent belief that illegal possession of firearms generally does not endanger public safety. Second amendment advocates should have a field day with this; how can anyone justify increasing gun regulation for the law abiding when it is now enshrined in policy that gun laws (at least in some cases) will not be enforced for criminals. What I wondered, however, is whether evidence supports the claim that moving police focus away from routine traffic stops will in fact reduce racial disparities in the criminal justice system. I looked at the 2019 arrests by offense and race from the Uniform Crime Reporting Program. The chart below shows the offenses sorted from highest to lowest by the percent of those arrested who are white. 


Note that 'property crime' and 'violent crime' are aggregates of other crimes listed; the data presented here is not focused on number of arrests, rather it is focused on the tendency of arrests for specific crimes to predominately affect white or non-white people. There may be some disparity between the way that law enforcement and the census bureau collect data about race, but according to the census 76.3% of the country was 'white alone' in 2019, implying that white people are overrepresented in arrests for just three types of offenses: driving under the influence, liquor laws, and suspicion. You may be wondering what 'suspicion' is. The FBI defines it as “Arrested for no specific offense and released without formal charges being placed.” Out of arrests for all crimes in 2019, 69.4% of those arrested were white. Nine types of offenses exceeded 69.4% for white arrests. 




So how is John Choi's plan likely to impact the racial breakdown of who gets prosecuted? Weapons offenses rank fourth from the bottom (55.6% white), so dropping these charges may reduce racial disparity in prosecution. But drug arrests rank sixth from the top (71.2% white), so a shift away from drug enforcement is unlikely to have much impact on the disparity. It strikes me that the crime most heavily weighted toward white arrests, driving under the influence, would probably be a common arrest during routine traffic stops. Isn't an open alcoholic beverage container or slurred speech one of the most likely pieces of incriminating evidence police might notice while looking through the window of a pulled over car? At the other end of the list, robbery and murder are the crimes least likely to involve the arrest of a white person, and by a large margin: 44.7% and 45.8% respectively. I suppose police might sometimes notice items that had been reported as stolen in a pulled over car, but I think this would be somewhat rare. If police find a dead body during a routine traffic stop, I would think that Choi would still carry out the prosecution. I guess I could be wrong.


One might question whether Choi is stepping outside his responsibilities. People sometimes accuse judges of 'legislating from the bench'; is Choi legislating from the attorney's office? Granted, there are probably many examples of elected leaders trying to influence policy in other areas of government, and discretion in prosecution is nothing new. In this case, however, Choi's concern apparently is not that those accused don't deserve to be prosecuted, but rather that he wants to persuade police departments to modify their policies on routine traffic stops. Aside from the question as to whether elected city leaders should be making decisions about how to run their police departments rather than Choi, I think there is a question about who suffers as a result of Choi's lesson to police departments. Assuming that prosecution really does reduce crime, Isn't it the general public that suffers for the sake of Choi's use of the attorney's office to manage police departments?