Total car sales are 30% below where they were four years ago before the Global Financial Crisis. We investigate whether car sales will return to their former level within the foreseeable future.
This article is an update of a piece written by my colleague Nigel Pinkerton. The article provides an estimate of average annual total car demand over the next five years, using three fundamental drivers of car demand: replacement demand, the ownership rate, and household formation. Based on this analysis, it appears that car sales are currently well below fundamentals, suggesting that demand pressure is building.
However, results are sensitive to a change in the ownership rate of cars per household and the scrap rate. Therefore attention must be paid to:
- factors that influence the discretionary element of car demand
- increases in the average age of cars when scrapped.
Replacement demand is a large source of car demand. In this article, replacement demand is measured by the scrap rate, which is the number of cars consigned to the scrap heap each year as a percentage of the existing fleet.
Replacement demand is primarily influenced by two factors: the age distribution of vehicles in the car fleet, and the underlying state of the economy. Replacement demand should increase as the proportion of older cars in the fleet rises because many of these vehicles become uneconomic to maintain. More cars will also be replaced during periods of strong economic growth as consumers are more confident about making big purchases and credit is typically easier to access.
Graph 1 depicts the scrap rate over recent history alongside GDP growth2. It is evident from this graph that the scrap rate moves with GDP growth quite closely. Between 2000 and 2008, when economic growth was strong, the scrap rate averaged 7.5%pa, whereas over the last three years of weaker growth the scrap rate has averaged only 5.5%pa.
The average age of cars when scrapped in 2010 was 18.02 years, up from 17.42 years in 2007. This increase was driven by used imports, whose average scrappage age increased from 16.34 to 17.51 years. The scrappage age of New Zealand new cars over the same period remained remarkably stable, hovering in the 18.7 to 18.8 year range.
The increase in the average scrappage age of used imports between 2007 and 2010 was not only due to buyers deferring the replacement of their cars during the recession, but was also due to the age distribution of the imported fleet. The used import car market did not open up significantly until the mid-1990s so few older used imports exist. Furthermore, frontal impact standards, introduced in 2001, led to a spike in imported vehicles that were manufactured around 1996.
The outlook for the scrap rate over the next five years will be driven primarily by the spike in used imports manufactured in the mid-1990s. In 2010, 17% of the used import car fleet was at least 19 years old, but by the end of 2015 the proportion of the 2010 fleet that will be in this age bracket will have risen to 64%. This glut of older cars will increase the average scrappage age of imported vehicles over the next five years.
However, an upper limit to the scrappage age must exist as eventually cars become too unreliable to continue using. For the purpose of this article we assume that the average scrappage age of used imports will rise to that of New Zealand new cars. We have rounded this age up to 19 years.
In volume terms an additional 624,784 of used imports will be at least 19 years old by the end of 2015. If one also adds to this figure the number of New Zealand new cars that will reach this age, a total of 838,254 additional vehicles will be at least 19 years old by the end of 2015. If all of these cars were scrapped and replaced over the next five years, then replacement demand would average 168,000pa. Assuming that the total size of the car fleet doesn’t change, this total equates to a scrap rate of 6.5%pa. This figure is entirely reasonable compared with past rates and fits with our expectation of a continued modest recovery in GDP growth.
It is worth noting that the state of economic activity and household confidence may serve to dampen or boost the assumed scrap rate. Given the current uncertainty surrounding the European sovereign debt crisis, many people will continue reducing debt for now and will delay replacing their car. However, cars eventually need to be scrapped, so any delays should just lead to more replacement demand in later years.
The above analysis assumes that the size of the car fleet will remain unchanged. However, in reality the fleet size will change. Two factors that may drive changes to the fleet size are car ownership rates, and household formation (to be discussed later).
New Zealand’s car ownership rate was 1.41 cars per household in September 2011, down from its all-time high of 1.47 in December 2007. Rising wealth and affordable prices for used imports have seen the car ownership rate climb substantially since the early 1990s, when it averaged around 1.20 cars per household.
Car ownership rates fluctuate because car ownership is a function of both need and want. The "need" component of demand is largely driven by factors such as the need to transport one’s children, go shopping, and get to work. Often there is little scope to completely substitute this demand component for other means of transport, so it should remain reasonably stable.
Much of the volatility in car ownership rates can be explained by the "want" component of car demand, which is discretionary by nature. This discretionary component of demand will increase when wealth is rising, wage growth is high, and car prices are low. However discretionary demand falls when credit conditions tighten, or if unemployment, petrol prices, or vehicle running costs increase.
There is little chance of the ownership rate increasing in the next five years. Currently households are demonstrating a desire to reduce their debt burden, and we expect petrol prices to remain high. Improved public transport in some regions will also offer alternative transportation options.
However, the ownership rate is unlikely to fall substantially either. Ownership rates have exceeded 1.40 cars per household for much of the last decade, and household routines have become accustomed to this level of access to vehicles. A continuation of this behaviour will be supported by our expectation that car prices will remain relatively favourable.
For the purpose of estimating underlying car demand, we assume an average of 1.40 cars per household over the next five years (although we have also undertaken some sensitivity analysis using lower ownership rates).
The number of households also influences the size of the New Zealand car fleet. Over the five years to March 2011, the number of households in New Zealand grew by 6.3%, and we forecast the number of households to grow by 5.6% over the next five years.
An estimate of underlying car demand
In order to estimate underlying demand for cars, we must combine our assumptions regarding replacement demand, the car ownership rate, and household formation.
Firstly, combining the assumptions of an ownership rate of 1.40 cars per household and household formation of 5.6% (over the next five years) suggests that the car fleet will grow by 88,400 over the five years to March 2016. This represents a net expansion of the size of the New Zealand car fleet of 16,700 each year on average. Demand from new households more than offsets the expected easing in ownership rates.
Once replacement demand is brought into the equation (using an assumed scrap rate of 6.5%pa of the current fleet), we expect total underlying car demand to average 169,900pa over the next five years. This estimate is more than 20% above the current annual running total of cars being registered for the first time.
Risks to the central view
The main risk to our estimate of underlying car demand is if the car ownership rate falls. As a result, firms in the car industry should pay close attention to factors which drive both discretionary car demand, and substitution away from car usage. To reflect this risk we estimated scenarios where the ownership rate falls to 1.35 cars per household (a level last seen around 2002) and where the ownership rate falls to 1.30 cars per household (last seen in 1998). The results are shown in Table 1.
Another risk arises from the potential for a permanent increase in the average scrappage age of vehicles. This could occur, for example, as a result of improvements to rust prevention technology in the early 1990s having extended the life expectancy of cars. To reflect this risk we estimated a scenario where, on average, cars are not scrapped until 21 years of age. Under this scenario, and given the current age distribution of the car fleet, this would equate to a scrap rate of 5.2%pa of the current fleet over the next five years.
As Table 1 shows, different assumptions about the ownership and scrap rates could potentially result in significantly weaker underlying demand than our base case of 169,900pa over the next five years.
The composition of car demand
Our baseline estimate of underlying demand for cars of 169,900pa over the next five years does not take into consideration how the composition of first-time car registrations could change. When purchasing a vehicle, buyers must not only choose between a NZ-new car and a used import, but also if the car they purchase is large (bigger than 1600cc) or small.
As of 1 January 2012 all imported Japanese cars must meet Japan’s 2005 exhaust emission standards. This change will effectively limit the age of imported cars from Japan to a maximum of six years old. The lack of older used imports will benefit dealers of new cars, as the price differential between a new car and a newly imported used car will be less.
Over time the pool of complying used vehicles will grow. However, the possibility of further regulatory tightening remains, and competition with buyers from other nations for used Japanese cars will put pressure on prices. As a result new car sales may form a larger proportion of total sales over the next five years than the 44% they represent at present.
Small cars (no bigger than 1600cc) currently make up 30% of total car sales, but this share is likely to rise over the next five years. We expect the appeal of small cars to buyers to increase due to high fuel prices, increased environmental awareness, lower depreciation expenses, and the reduced availability of inexpensive older used imports. Car manufacturers will support this trend by continuing to expand their range of smaller cars and including features that have historically been reserved for larger alternatives.
2 GDP growth shown in the graph is lagged four quarters, due to a quirk in car fleet statistics data. Unless a vehicle has been actively de-registered, NZTA does not exclude it from fleet statistics until 12 months have passed since it was last registered. In other words, it takes up to a year for a scrapped car to appear in the data.