07 Aug How to lie with graphs

PC Leader Jamie Baillie’s election promise to hold power rates at current levels came in a position paper that included the following unsourced graph, purporting to show that something called “energy costs to rate payers,” measured in units it did not explain, have increased by 27 percent since 2009:

Wow, that certainly looks shocking!

Contrarian is no statistician, and my graphic skills are tenuous, but I read Darrell Huff‘s classic How to Lie with Statistics shortly after it came out in 1954, and Chapter 5, “The Gee Whiz Graph,” stuck with me. Of the persuasive power of graphs, Huff had this advice for readers deploying them to “win an argument, shock a reader, move him into action, sell him something:”

Chop off the bottom.

Baillie did exactly that with his “energy costs” graph. He chopped off the bottom 64% of the graph. If you start the y-axis at zero, and display the graph in the same horizontal format, it looks like this (with apologies to readers skilled in PhotoShop and Illustrator).

It’s still a significant increase, but not quite so scary or election-worthy as Baillie’s manipulated format. If anyone has the time to parse exactly what’s included in, and excluded from, “energy costs to rate payers,” I suspect we will find that Baillie has selected the fastest rising component of electricity bills to inflate his point.

Bear in mind, too, that the period covered by the graph roughly corresponds to a reduction in Nova Scotia Power’s use of dirty coal from 80 percent to just above 50 percent. That phenomenal drop is a good thing, rarely mentioned by the company’s critics.

Even so, it’s reasonable to ask whether there is some mechanism that could give Nova Scotia access to North American electricity markets and the pricing stability they could bring. Since the first electricity was produced in Nova Scotia early in the 20th Century, we have never had the ability to import or export significant amounts of power.

The reasonable answer is: The Maritime Link, whose principal side benefit, steadfastly ignored by critics, is the robust connection to the North American Grid it would create through Newfoundland, Labrador, and Quebec to the north, New Brunswick and New England to the west. Aside from a guaranteed 35-year supply of predictably priced power, that is the best argument for building The Maritime Link.

[Disclosure: From time to time, I have consulted for NS Power and Emera on issues related to power rates and the Maritime Link. Reasonable people can and do disagree about Nova Scotia energy issues, but they ought to avoid misleading graphics.]