Trying to understand state excise taxes on beer.

The Pub
1

I saw this posted the other day:

A slight correction of the numbers are here (from the Beer Institute):
https://www.documentcloud.org/documents/325601-excisetaxrate.html

I was wondering what you guys thought might make these tax rates vary by state? Let me tell you that I punched these numbers into SPSS (stats software), used OLS regression, and found that %Christian, %liberal, perCapita car fatalities, property tax rate, and #breweries per state all had no effect.  It was all state level data.  I am leaving out of this explanation the one and only variable that had a coeeficient that was significant (it is positive and a can of worms).  I am thinking of writing this up, but am interested in including more “controls” or explanations that help us understand state excise tax policy.  What do you think the model is missing?  I should probably include medianIncome and maybe percapita # of breweries in the state.

2

I don’t think there’s much mystery. States with breweries big enough to do substantial lobbying (CO, TX, MO, WI, PA) have low taxes.

3

Big breweries might matter but #of breweries doesn’t.  Maybe I should get a figure for the number of brewery employees for each state.  The more people are employed by breweries then the lower the taxes on beer, right?

4

^ Nailed it. Except Wyoming, they just don’t really have taxes there.

5

Or people from what I remember.  Population density is the next to lowest in the country, next to Alaska.  Western Wyoming is beautiful.

6

Alright.  So if you exempt those 5 states, what explains the rest?

7

ABInbev also have breweries in CA, GA, VA, NY, NH, FL, VA.

MillerCoors also in CA, GA, NC, VA.

Not all are low tax states. Just saying.

8

Well, I’m not going to run the regressions, but it sounds like you have, so you tell us.

This might help: http://alcoholjustice.org/campaigns/charge-for-harm/450-neglected-and-outdated-state-beer-taxes.html

Maybe legislators in some states are just lazy?

9

Fair point, but if you except the South, where they’re just into that kind of thing, it still applies.

10

I missed NJ. You are probably right. Big business often has many reasons to locate a facility, taxes are just one.

I am wondering what the can of worms correlation is.

11

GA and AL have a “uniform statewide local excise tax.” I have no idea what that means, but the state excise tax for GA is actually $0.48 and AL is $0.53. And Wyoming passed their tax back in 1935.

The rate in that graphic for WA is actually the old “high tax” rate. The new 2010 WA “high tax” rate for beer is $0.76, and the “low tax” rate for small and domestic breweries stayed the same at $0.15. I haven’t checked the other numbers in the graphic, but at least that one is wrong.

12

I was just running the numbers yesterday and according to my analysis the only variable that had a significant coefficient was %populationBlack.  According to the regression model, as %Black increases so does tax on beer (about .40 cents for every percentage point).

I told you it was a can of worms. This finding is, however, consistent with some of the prohibition literature I have read (check out Okrent’s Last Call).  This also seems analogous with some of the the political science (my day job) research in other policy areas.  I suppose if I include other variables like median income the relationship between race and beer tax might disappear.

13

what’s the r^2 for that correlation?

14

The model explained about 17% (r-square) of the variation.  It is a small r-square.

15

So 0.17? That’s an incredibly small coefficient of determination.

16

Not really.  Not to get into the methods stuff but the r-square isn’t as important as the actual coefficients presented in the model.  The r-square is just a descriptive statistic that describes the model’s goodness of fit. To increase its size all one has to do is to include a bunch of variables.  Check out “How not to lie with statistics” by Gary “The Political Science Methods God” King: How Not to Lie Without Statistics | GARY KING

But I agree, it’d be nice to have a larger R2.  What other variables would you include?

Thanks for your responses.

17

I think you need to use better data, and all the data should be consistent. Many states have different excise tax rates for large vs small breweries, like WA’s high and low. Sure, GA has uniform “local” excise taxes, but what about all the other places in the country? What if all of the breweries in some other state, are in one location that does have a local excise tax?

I still think 0.17 is worthless. Or let me put it this way: I would not wager any of my own money on those odds. Sure, you can add a bunch of co-linear variables to inflate your R^2, but I think it’s safe to say % of black residents is not related to the excise tax rates you’re using.

18

See my Sig…

19

I love my country, and have no problem paying taxes to help make it the greatest place in the world. After all, you get what you pay for in life. I really don’t get why some people hate taxes.

Case in point, in rural MO there is no dog catcher, and there are dozens of packs of wild dogs that run around, eating pets and harassing livestock. Given the large number of well-armed people here, this shouldn’t be a problem, but no one bothers to shoot them. In CO taxes were way higher, and it was a lot nicer to live there.

20

No matter how I try I cannot reply without entering the verboden zone :cry:
But there was something about the revolutionary war when colonists revolted
from England due to taxation…or something like that if my recollection of the
HISTORY that I was taught is correct…

Mods feel free to delete if it is too far…in the realm of the “p” word