Best Correlation-Causation-Confusion Quote

This is by far the best quote on the fallacy of confusing correlation with causation. This comes from my daughter’s second favorite author Lemony Snicket (Daniel Handler)

If you work hard, and become successful, it does not necessarily mean you are successful because you worked hard, just as if you are tall with long hair it doesn’t mean you would be a midget if you were bald.

This quote, in my mind, displaces the previous best quote by my favorite economics author Greg Mankiw (not my favorite economist however)

Suppose we were to graph average SAT scores by the number of bathrooms a student has in his or her family home. That curve would also likely slope upward. (After all, people with more money buy larger homes with more bathrooms.) But it would be a mistake to conclude that installing an extra toilet raises yours kids’ SAT scores.

Now back to reading, “What can we learn from …” articles I have Instapaper-ed

Watered Down Profits – Shifting Demand Curves

I saw a commercial for PUR water purifier that makes a value proposition on savings from bottled water. There was another commercial by Wal Mart that showed people buying their private label bottled water and taking it with their brown bag lunch. There is yet another commercial by Brita that takes a green approach by showcasing the ill effects of plastic bottles. The net is bottled water sales are down due to a variety of reasons.

  1. How should the marketers handle the shift in consumer preference?
  2. Should they drive down prices to keep their customers from switching to tap water?
  3. How much will the sales volume increase with price change? i.e., what is the price elasticity of demand?
  4. Is raising prices an option?

It is almost out of luck there is data on all these (from WSJ) that one can find without searching too hard. It is as if do all of us a favor, one of the biggest brands, Coke, did not change its prices and another biggest brand, Pepsi, lowered its prices. For all practical purposes we can treat that there is no differentiation between these brand for bottled water and they represent the overall market.

Coke, kept its prices stead and saw its sales drop by 26%. Note that this is sales in dollars not units, but at constant prices we can assume that this represents volume drop. I am no Greg Mankiw or Andy Rose, but to me this drop in demand at constant prices seems like  a shift in the demand curve. So let us treat this as the shift in the demand curve due to income effect (people switching to tap water or store brands as part of their cost cutting). See Figure 1 and 2 for the shift in demand curve.

Figure-1 Demand Curve
Figure-2 Demand Curve Shift

Figure-1 were the demand curve before recession, Figure-2 shows the shift. It is an approximation and not an accurate drawing.

Next, Pepsi cut its prices by 5%, and its sales (dollars) fell only by 13.8%. This is shown in Figure-3. But that is based on the previous demand curve. Since we assumed the demand shifted down, the 13.8% drop is actually 12.2% (26% – 13.8% = 12.2%) increase in volume. That is the price elasticity of demand on the new curve at the same price point before the price cut is 12.2/5 = 2.44.

Figure-3 Price drop results in sales increase (along shifted demand curve)

Not all of this 12.2% sales increase translate into profit. Pepsi and Coke do not break out their earnings to show revenues and profit from bottled water. Luckily, Nestle does. If use their numbers EBDITA numbers, the margin as a percentage of sales is 7.3%. This seems on the lower side. In 2003, MorganStanley reported that the margin is around 20% for US bottlers. So we can take the margin to be in-between these two estimates, say 15%. So Pepsi earned a gross profit of  12.2%*15% = 1.83% of the total sales.

Not a bad move, better than Coke which did not change its prices. But could they have done better? What if they had raised priced by 5%? That is a 5% of sales added to profit.  But their sales volume would fall. If we take the price elasticity of demand to be the same (2.44) in the other direction, then their volume would have fallen 12.2% with a profit loss of 12.2% * 15% = 1.83%. This translates  to a incremental gross profit of 2.6% from price increase despite a 12.2% drop in sales. If Coke or Pepsi were to reduce capacity and operational expenses to account for this 38.2% (26%+12.2%) drop in volume, the cost savings will add to the gross profit from price increase for a higher operational profit.

This leads us to conclude, that  price increase at lower volume would have delivered higher profits than sales increase from lower prices. But recommending price increase when the sales just fell 26% is going against the “conventional wisdom” – one needs data, conviction to act on it and limitless courage to go against “conventional wisdom”.