TCS Daily

The Good News on GDP May Be No News At All

By Kevin Hassett - May 3, 2001 12:00 AM

We found out last week that GDP rose 2 percent in the first quarter, well above expectations. Markets soared in response to the good news. With less fear of an impending recession, investors decided that profits and equity prices have bottomed out.

Was the news that good?

Probably not. Indeed, there was almost no news in the GDP report. Here's why. The Commerce Department's Bureau of Economic Analysis (BEA) calculates GDP by adding together data items that come from a number of sources. Business investment, for example, is based on the data on orders and shipments of durable goods. The BEA calculates the business fixed investment component of GDP by adding shipments for each month in the quarter together, and then adjusting this sum with trade data.

So why the big surprise? The GDP number is often released before the BEA has complete data on all of the components. In other words, the BEA is forced to make assumptions about the missing data. The GDP number came in strong because the BEA assumed things were better than most economists believe in the unmeasured months. My belief is that things were not that good, and that the first quarter will look a whole lot worse when we receive the final data.

Thus far, we only have data on inventories through February. My guess is that March is going to look much worse than the BEA assumed. To see why, allow me a little digression.

If you run a business you probably spend a great deal of time managing your inventories. Profits can disappear overnight if you mess it up. If consumers are lining up to buy your product, then you better have tons on hand to avoid delays. If consumers decide to take their business elsewhere, then a warehouse full of unsold merchandise will cost you dearly. Since you have to make a product well ahead of its sale, production always takes place under a cloud of uncertainty. Will buyers be there when your product hits the shelves?

Back in the late 1940s, economists first noticed that business inventories tend to swing wildly around recessions. Moses Abramowitz, for example, wrote in 1950 that U.S. recessions prior to that time had been periods of sharp inventory liquidations. These liquidations were of an enormous scale, often explaining much of the decline in GDP during recessions. This suggested that recessions happen in part because firms predict the future poorly and get stuck holding the merchandise.

Up until the 1990s, this pattern stayed the same. In good times, firms would gradually ramp up production in response to booming consumer demand. Then one day consumers would disappear suddenly (and without notice). Firms at first would interpret the slack demand as a temporary phenomenon and maintain healthy production, but in recessions consumers continued to stay away. Output would pile up in warehouses, and eventually firms would have to dismiss workers and liquidate inventory in order to stay in business. Factory doors would bolt shut, and stay locked until inventories had been worked down to acceptable levels.

Lots of things go south in a recession, but inventories are the undisputed champions. In a typical postwar recession, the decline in inventories accounted for about half of the overall decline in GDP! This was true in the 1991 recession as well, when inventory declines accounted for 49 percent of the drop in GDP.

Which brings us to the story that led many new economy theorists to be hopeful that recessions would be a thing of the past. In the 1990s, firms began to use computers to improve inventory management in two ways. First, information about demand is now coordinated and transmitted over the Internet. Upticks in the aisles at Home Depot send immediate signals to factories that more production is needed. Second, computers have allowed firms to adjust the type of goods that they produce in order to make sure that hot items are in stock and unpopular items are not. Computers allow production to turn on a dime. With these changes, many hoped that the wild inventory swings would be a thing of the past.

Guess what. They are not. Business inventories in the U.S. dropped fairly sharply in February after massive accumulation in previous months. Especially disturbing was the inability of firms to get ahead of the curve. While inventories dropped, the ratio of inventories to sales increased because sales dropped more. This is ominous.

Firms set their inventory targets based upon how much they think they can sell in a month. When their inventory-to-sales ratio increases in a downturn, it usually sets off fire sales. So the March numbers will likely be ugly. If they are, we will ultimately look back on the first quarter of 2001 as being far worse than we originally thought.

The key will be what happens next. Will the inventory-to-sales ratio return to a reasonable level? Will declines in sales run ahead of the inventory correction again, pushing us closer to the recession abyss? We will find out on May 14th when we get the March inventory data.

TCS Daily Archives