Pub. 58 2017-2018 Issue 1

35 FALL 2017 This conclusion leaves us to base our 2025 USA new-car sales volume prediction on more traditional inputs (such as population and income growth, number of drivers on the road, etc.). Not a lot of forecasters look that far ahead, but we can point to two who have tried: • IHS uses a fairly straightforward approach, looking at GDP, population, and other factors, and comes up with 17.3million sales in 2025, essentially unchanged from 2015’s 17.5. Their forecast implies a declining number of cars demanded per capita, since the Census Bureau predicts an 8%growth in population by then (half from natural increase and half from immigration). This makes sense, as the American market is fairly saturated: as Experian points out, the average American household already has c. 2.1 cars and light trucks (260 mm vehicles and 125 mm households). Put another way, we have 1.2 vehicles for every licensed driver in the USA (215 million), so we can’t even drive what we have now – at least, not all at the same time! 6 • The Michigan-based Center for Automotive Research (CAR) took a somewhat different approach and came up with a baseline SAAR for 2025 of 17.9 million, a bit above today’s rate. CAR tried to explicitly address price changes, since obviously as car prices rise (adjusted for inflation), demand should fall – and most analysts and OEMs expect car prices to continue to meet or exceed inflation, as we add more content to vehicles. Some of this content is unregulated (e.g. more speakers for the sound system), but much is regulated (e.g. more hybridized powertrains, to meet tightening MPG and GHG 7 regulations). This kind of forecasting is very difficult to do, because content-driven price increases cut both ways. For example, if I have to spend $2,000 more on a car because it has higher mandated MPG (persuading me to hold off buying), I will also save money on the gas I don’t have to burn (inducing me to accelerate my purchase). But in CAR’s model a 10% increase in the inflation-adjusted net price of cars (netted across costs and benefits), which is not unreasonable, could knock 2.5 million units off SAAR in 2025. 8 So we will all have to keep an eye on vehicle costs. We won’t pretend to have better automotive sales forecasting skills than IHS and CAR, so we’ll use as our baseline forecast the average of the two, or 17.6, which is just about where we are now. 9 One more comment on the SAAR in 2025: it is very likely to be skewed to the richer end of the product mix, and to richer customers, than it is today. This is because with every year car quality gets better, and so used cars last longer and longer: in 1975 the average age of the light-vehicle fleet was about 5.5 years, and today it is around 11, just about twice as long. 10 With more and more used cars on the road, of higher and higher reliability, they become better deals relative to new cars. The era when a person bought a used car because she just couldn’t afford a new one is over: today the used car is, in terms of pure economics, usually a very sensible option. So cars have become more like houses: the wealthy (on average) buy new houses, and the less-wealthy (on average) buy used ones. 11 As evidence on this point, note that the average transaction price of new vehicles has slightly outpaced overall CPI (Consumer Price Index) inflation ever since the Great Recession. We can see this shift already taking place: in 1973, for example, when we sold 14 mm new light vehicles, our population was 211 million, versus 17.5 mm in sales and 320 mm in population in 2015. So in 1973 one in 15 Americans bought a new car, but in 2015 only one in 18, a fraction which we can assume comprises customers of higher incomes. This move of the market to the wealthier segment of the population hasn’t gone unnoticed by OEMs: note the “democratization of luxury” initiatives by companies likeMercedes, as they produce more (relatively) down- market models like the CLA andGLA. Dealers should therefore expect to steadily upgrade processes, people, and facilities over the next decade, to meet the more exacting demands of an ever-wealthier customer base. SPECIAL TOPIC Chapter A: Rural Futures Throughout this report we’ll be talking about some generic “average” dealership, maybe a mid-sized suburban facility—and our findings broadly apply to this store. However, there is another category of store which requires special, separate treatment, because of its different circumstances, and this is the small rural store, typically family-owned. These smaller stores make up a significant portion of NADA membership: about 40% of the Association’s membership retail 300 or fewer new vehicles each year. And they do deserve separate treatment, because the circumstances they face are different. We know this in part from looking at other retailing sectors: we don’t seemassiveWalmarts inmidtownManhattan, and we don’t see Starbuck’s in remote farming communities: different markets require different retailing approaches. Accordingly, in researching this report we paid special attention to the situation facing smaller stores, and put on more than a few miles driving in lesser-populated states, in order to see first-hand  DEALERSHIP OF TOMORROW — CONTINUED FROM PAGE 33  DEALERSHIP OF TOMORROW — CONTINUED ON PAGE 37 6 See periodic IHS Automotive forecasts by IHS Markit, US Census Bureau projections, and periodic Automotive Quarterly Briefing decks from Experian Information Solutions. 7 GreenHouse Gasses, e.g. carbon dioxide 8 “The U.S. Automotive Market and Industry in 2025,” from CAR, Ann Arbor, Michigan. 9 One reason to like this 17.6 number is that it adheres to the famous Persistence Method of Forecasting, from meteorology: “The best predictor of the weather for any particular tomorrow is today’s weather.” Believe it or not, this method has a very good track record. Weathermen hate to acknowledge this, for the obvious reason. 10 See various fleet data press releases over the years, from R. L. Polk. 11 If the car/house analogy seems far-fetched, note that the median age of a single-family home in the USA is only 35 years, according to the most recent American Housing Survey (sponsored by the Department of Housing and Urban Development (HUD) and conducted by the U.S. Census Bureau).