It is difficult to make predictions, especially about the future
In the news business, there’s an adage that says be the first to get the story, but first get it right. NRF isn’t the only organization that issues a holiday spending forecast, so there has always been pressure to get it first in our industry as well. There has been equal pressure to get the forecast right, and we have succeeded more often than not. But this year, both the pressure to get it right and the difficulty of doing so have been greater than ever. And that’s why we chose to put carefulness and caution ahead of speed by moving our holiday forecast from its traditional timing in the first week of October to the week of Thanksgiving.
Rather than release a forecast at a point when data would reflect retail sales only up through August and not tell us enough about what to expect in this unusual year, we decided to wait, and minimize oversights in the outlook. Releasing our forecast later allowed us to take October sales data and other recent indicators into account. We don’t claim holiday sales will match our forecast to the penny — economists have to rely on assumptions and even with more data the forecast is only a range of possibilities rather than a precise target. The goal of waiting was to ensure a more reliable forecast that provides useful information for those who rely on it for decision-making.
This year, NRF is forecasting that holiday retail sales will grow between 3.6 percent and 5.2 percent over 2019. That works out to between $755.3 billion and $766.7 billion. By comparison, last year holiday sales grew 4 percent from 2018 and totaled $729.1 billion. Holiday sales have increased an average 3.5 percent over the past five years. Our forecast includes online and other non-store sales, which we expect to increase between 20 percent and 30 percent to a total of between $202.5 billion and $218.4 billion, driven by this year’s strong shift to online shopping. That is up from $168.7 billion last year.
Let me explain the mechanics of the forecast. As always, our modeling begins with current data and then makes assumptions about each economic indicator’s level, direction and rate of change — all adding up to insights on consumers’ ability to spend and their willingness to do so. We look at key variables such as employment, disposable income, past retail sales and consumer credit, among others. This year, in addition to the usual array of monthly and quarterly data on these issues, we have taken into consideration “high-frequency” data that comes out as often as weekly and provides more up-to-date snapshots of consumer mobility and the current state of the economy. In addition, this year’s forecast required us to consider unique factors such as whether Congress would provide additional fiscal stimulus and the unsettling impact and timing of the recent widespread coronavirus surge. The virus is undoubtedly the largest uncertainty and biggest risk the economy is dealing with.
This framework lets us come up with the forecast of what we believe has the greatest likelihood of occurring. But in this unusual environment, we must recognize that the final results could come in higher than predicted, particularly in the event of spending exuberance among households wanting to have a better-than-normal holiday to compensate for the difficulty of the last 10 months. On the other hand, spending could be lower, depending on what shapes the economy, labor market and consumer confidence. There is no doubt this is the most unusual economic environment in our lifetimes. Given the pandemic, there is significant uncertainty regarding both consumers’ ability to spend and their willingness to spend. I cannot think of a period with so many simultaneous factors hitting the economy at once when formulating this forecast.
There are several factors we know will come together and have a major impact on the outlook, and some of them are unusual for an economy moving out of a recession and into a recovery period. Many consumers have the ability to spend this season given better-than-expected job and wage gains and strong financial markets. Employment has increased for six consecutive months and the jobless rate has fallen to 6.9 percent — almost double pre-pandemic levels but less than half the peak seen in April. There have been record savings, accumulated from a combination of unspent government stimulus and enhanced unemployment benefits along with reduced spending on personal services, travel, entertainment and dining out. Energy prices are lower than a year ago — gas at the pump is down along with winter heating oil and natural gas — and rapidly rising home and stock values have also created a tailwind for consumer spending. All of this has been evident in retail sales growth of 6.4 percent in the first 10 months of 2020 compared with the same period a year ago. Nonetheless, in contrast to those positive factors, there are households that have not benefited from the labor market or financial markets and are finding it difficult to buffer the health and economic disruption of the pandemic.
There are other factors that might not be thought of as economic indicators but still have an impact. Weather always plays a role in holiday retail sales, and while details vary by region, the National Oceanic and Atmospheric Administration's summary favors cooler, wetter weather in the north and warmer, drier weather in the south. This climate phenomenon has correlated with stronger retail holiday spending in the past and could be a factor in 2020.
The near-term concern is the long shadow cast on the economy by the surging virus and expiring government support. Every virus indicator across the United States is elevated and accelerated, which could pump the brakes on the momentum we have seen and have consequences for spending. Consequently, the outlook comes with significant uncertainty and risks. We are possibly looking at a much higher number of cases in December due to the large gatherings of friends and family that took place during the Thanksgiving holiday. Imposing new restrictions on mobility and business operations in response could drain consumer confidence and spending.
This is a year when it is particularly important to recognize known risks, understand unusual factors and attempt to reduce oversights when making predictions. Given that, we have put an extra-thorough process of examining the complex and sometimes-conflicting data before us ahead of being first. Despite uncertainty, we are optimistic that households will keep up the energy of spending through the end of the year. Like any other forecast, it is part science, part intuition and part luck. As the holiday season progresses, we will continue to closely watch data and developments for guidance.