It’s Shocking

by MaxPower

Recently, as I was scanning the newswires I came across an interesting header from the British Columbia Automobile Association (BCAA) on July 28th: “High gas prices affecting motorist’ driving habits“. While not an intuitively a shocking statement, I personally found it interesting as I had done an econometric study, as my undergraduate economics thesis, which concluded the cost of gasoline has virtually no impact on the amount people drive or their propensity to use public transit. Was I wrong or is the press release and the “survey” the BCAA conducted wrong? Guess.

In doing my study I looked at the amount and price of gas sold in Alberta over a 30-year time frame, the number of cars on the road and their average fuel efficiency (over the 30 years there were more cars on the road but with better fuel efficiency so that total gas consumption stayed relatively constant), the number of public transit riders and deduced per capita fuel consumption. My conclusion was that gasoline was very inelastic (demand varying little with changes in price) and consumption patterns would likely not change materially until a price shock of some magnitude.

Of course, this press release was trying to give the BCAA a little bit of free press by playing on the fact that gas prices are ‘too high’, even though it has been proved through multiple studies that Canadians pay some of the lowest gas prices in the world. Here is a nice page (http://www.energy.ca.gov/) from the California Energy Commission, which shows out of the largest developed economies in the world Canada has some of the lowest gasoline prices, second only to the US. British gasoline prices by comparison are 2.5 times higher than average Canadian gasoline prices. Or perhaps you prefer the CBC’s version of essentially the same facts: http://www.cbc.ca/news/background/gasprices/ . Regardless, even given the fact that gasoline prices aren’t ‘too high’ the BCAA decided they needed to do a survey on whether people’s driving habits were changed by these ‘high’ prices.

Note that the BCAA conducted a ‘survey’ that asks people for their opinions on a variety of composed question and is of much less statistical worth than a ‘study’, which is constructed without bias and corrects for sampling and reporting errors. The BCAA survey found that “50% of BC drivers have reduced the amount of driving they do in an average week to compensate for higher fuel prices”. Shocking. Especially considering my results, which show that getting 50% of people to change their driving habits would require a cataclysmic rise in gas prices. Just as shocking, are the other survey results: 38% of drivers are changing future consumption habits by ‘considering’ the purchase of a more fuel efficient car and 27% have changed their summer vacation plans to reduce the amount of driving required.

Sorry little Johnny, no trip to see Grandma in Alberta, gas prices have risen by 15 – 20%. But wait, the BCAA results are accurate +/- 4.4%, 19 times out of 20! That must mean the survey is statistically significant. Well it would be if the survey were a study. However, since the survey is asking people for opinions and the BCAA makes no attempt to see if people’s opinions are reflected in ‘real’ data then the survey is garbage. Even with a statistically sound survey if you ask garbage questions you get garbage answers. Garbage in, garbage out as my econ prof used to say.

The problem with surveys is that people think about what answer the survey-giver wants to hear. If someone calls you up and asks “will high gas prices impact the amount you drive?” I would expect the vast majority would say “yes” regardless of whether they actually drive less or not. A much better way of getting the answer would be to look at per capita gasoline consumption so to eliminate human bias. Between 42.4 and 33.6% of respondents on this survey ‘claimed’ to be ‘considering’ a purchase of a more fuel-efficient car because gas prices are ‘too high’. Lets see here, we have a survey-taker’s bias to respond positively to questions so there will be an inclination for more people to say yes, then the question asks if they are ‘considering’ (which could mean anything) buying a more fuel efficient car, all based on underlying reasoning that is incontestably false – that gas prices are ‘too high’.

Who is the BCAA trying to kid here; this is a sham of a study and in no way relates to what is actually happening. Has the BCAA seen people driving less? Is per capita gas consumption in BC declining after the spike in gas prices as compared to what it was doing before? Has the BCAA isolated potential conflicting data points, which would artificially reduce the amount people drive – for example, adverse weather has been shown to decrease per capita gas consumption over a couple weeks or a season. Has the BCAA seen the number of fuel-efficient cars being purchased increase after the gas spike as opposed to before?

Of course the BCAA hasn’t. They haven’t because they relied on an anecdotal survey of 500 people who responded to biased questions rather than doing some data crunching to come up with real answers. The survey is a sham, the conclusion drawn by the BCAA patently false and you can bet some journalist is going to pick this up off the wires and run a story on how ‘consumers are driving less because of high fuel prices’.

Lastly, as the kicker, the press release makes the statement “It’s interesting to note the number of drivers who attribute high gas prices to taxes… even though taxes are not the reason for the recent price hike”. Oh really? Taxes aren’t the reason? They sure compound things. See this page (http://www.caa.ca/gasprice/pricing.html) from the CAA (parent company to the BCAA) . A quote from the page “On average, Canadian taxes represent 40 to 50 per cent of what you pay at the pumps” seems to indicate taxes are a big chunk of the gas price. And when the base price (crude oil) goes up, the amount of tax charged on the purchase goes up too. Shocking. Crude oil, refining and distribution makes up 40 cents out of the average price of every litre of gas in Canada, then tack on the “Federal Excise Tax”, some “Provincial Tax”, and a bit of “Transit Tax”. So now the price is now considerably higher. But wait, then slap on the GST and (if applicable) PST/HST on top of that total (yes, you are being taxed on a tax). True, the recent rise in crude oil costs has been the driver for gas prices to increase by around 20%, but if taxes account for 50% of the total gas cost and are added incrementally to the price of crude oil, they would account for at least 10% (or half) of the price rise. Has this press release been proofread?

The BCAA should be embarrassed and ashamed at the shoddy work in this press release. It is biased, based on a survey, which was flawed in its construction, and presents a wholly inaccurate picture. Yet some journalist who doesn’t want to think about the methodology or the inherent biases built into the survey will take this story and run with it. That’s shocking.

MaxPower