Economic forecasts vs. reality

If a man isn't willing to take some risk for his opinions, either his opinions are no good or he's no good.

Ezra Pound, poet

For this last column of 2018, I want to look back at some of my forecasts from a year ago and then compare these forecasts to what actually happened during the past year. My hope is to identify both the hits and the misses and then determine how this information can be useful in forecasting the growth rates for important end markets and economic indicators for the plastics industry in 2019.

I think this type of review, formal or informal, is a common practice among owners and managers in a competitive marketplace. And if you have been in business long enough, then you know this can be a humbling exercise. What is less common is taking the risk of sharing the results of this year-end ritual in a column in a prominent magazine. So I hope Ezra Pound is right.

Since 2018 is not yet officially over, we are still a few weeks away from having all the pertinent data for this year. Therefore, some of the numbers on the table are estimates and subject to revision. But barring anything extreme between now and New Year's, I think we have a pretty good idea about how the large indicators should perform.

The trend in housing starts and the other indicators from the residential construction sector are hugely important to the overall economy and the plastics industry. This is an area where I focus a lot of attention.

The upward trends in this data had been reliable for the previous years, and this continued in 2018. My forecast from a year ago called for a gain of 5 percent in the total number of new houses that would be started this year, and it looks like this will be very close to the mark.

A year ago, mortgage rates and the unemployment rate were at very low levels, and household incomes were rising. The home sales data for both new and existing homes were in a long uptrend. There are also some longer-term demographic trends that continually push demand for housing higher, such as a rising population. I concluded that the net effect of these trends would be a moderate acceleration in the housing starts data from a gain of 2.5 percent in 2017 to growth of 5 percent in 2018.

As we enter 2019, there have been some changes to a few of these trends. On the positive side, the unemployment rate remains quite low and household incomes are still rising. The longer-term demographic data remain supportive, especially in the South and West regions.

But the trends in the home sales data hit their cyclical peaks and have gradually declined for much of this year. One of the reasons for this is that mortgage rates have increased by about 1 percentage point in 2018, and the trend is upward. The problem of affordability is therefore becoming more of a risk to the outlook for this sector.

I expect mortgage rates will nudge higher, but this will be partially offset by the projected rise in household incomes. My current forecast for housing starts calls for another increase in 2019, but the rate of growth will decelerate to a moderate 3 percent.

Following two consecutive record years in 2015 and 2016, total motor vehicle sales cooled in 2017. The total sales data is a good indicator of overall market demand, but it includes purchases of vehicles imported into the U.S. For domestic suppliers of this industry, I tend to focus on the number of vehicles assembled in the United States.

After falling by more than 8 percent in 2017, the number of motor vehicles assembled in the United States declined only 1 percent in 2018. Following the large drop in 2017, I expected a further substantial decline in 2018. But the overall data from the motor vehicles sector held up better than I expected this year. The mostly flat trend in gasoline prices combined with incremental gains in fuel efficiency have pushed demand for SUVs ever higher while demand for sedans has diminished.

We are still past "peak auto," and I do not expect to hit a cyclical bottom for the foreseeable future. My latest forecast calls for a decline of 3 percent in the total number of motor vehicles assembled in the U.S. Historically, the data from this sector has been sensitive to rising interest rates, so this will continue to be a headwind in 2019. But this should be partially mitigated by rising wages and a relatively low price for gasoline.

Retail sales, excluding autos, increased by almost 5 percent — inflation-adjusted — in 2017 when compared with the previous year, and the graph was gradually trending higher. Real wage growth at that time was still below 2 percent per year, but the combination of a huge gain in the stock market, a strong rise in overall home values and lower tax rates were generating a salutary wealth effect. In other words, consumers were in a good mood. This prompted me to call for another annual gain of 5 percent for total retail sales in 2018. It looks like I was just a bit low on this one.

Wage gains are expected to continue to accelerate gradually this year, and overall home prices will also stay strong. But higher interest rates will slow growth in the interest-sensitive segments. Higher interest rates will also impede the gains in the stock market and home values. Consumers remain upbeat but not at quite the same level as last year. I expect the uptrend in retail sales to continue, but the growth rate will moderate a bit. A moderating uptrend will also prevail in the data for basic consumer goods as well as consumer packaged goods.