 |
Marshall J. Vest
Forecasting Project Director
March 1, 2008
Many forecasters have revised their forecasts to include a short and mild recession for the U.S. economy during the first half of 2008. Arizona’s economy began contracting last August as the downdraft from the collapsing homebuilding industry pulled the rest of the economy along for the ride. Normally, Arizona lags behind the nation at cycle peaks, and reaches bottom at about the same time. It may take Arizona longer to stabilize this time, due to the nature of this cycle. The good news is the downturn should be shallow and similar to the 2001 recession.
The Business Cycle Dating Committee of the National Bureau of Economic Research (NBER) defines recession as follows: “A recession is a significant decline in economic activity spread across the economy, lasting more than a few months, normally visible in real GDP, real income, employment, industrial production, and wholesale-retail sales. A recession begins just after the economy reaches a peak of activity and ends as the economy reaches its trough. Between trough and peak, the economy is in an expansion.”
So, in order to qualify as a recession, the economy has to decline (shrink, contract) a significant amount, last at least several months, and permeate throughout the economy. Depth, duration, and dispersion (3-Ds) are all required.
Since 1960, the U.S. economy has experienced seven recessions (Exhibit 1). On average, they last 11 months. The longest lasted 16 months (twice, 1974-75 and 1981-82), the shortest only six months (1980). The two most recent both measured eight months. As for real GDP, the largest decline was 3.1% during the 1974-75 recession. The smallest drop was 2001’s decline of only 0.4%. The average drop in real GDP over the seven downturns is 1.7%.
As 2007 came to and end, the U.S. economy may have entered a recession, but it will be several more months before NBER can say for sure. It appears from several measures that the U.S. economy is now contracting, but will it decline far enough or last long enough to qualify as a recession? Time will tell. It will likely be the end of this year before NBER identifies the peak month – and by that time the recession could be over! They waited until November 26, 2001 to officially identify January 2001 as the peak month. Then on July 17, 2003, they announced that the recession had ended in November 2001. So, at this point, use of the “R” word is just a forecast.
Recent economic reports for the nation have been weak and decidedly downbeat in January. Vehicle sales were sharply lower and chain store sales were the worst in the history of the series, dating back into the 1970s. Employment registered its first decline, 17,000, since the last recession. The stock market continued to sell off and has now lost enough to be officially designated as a bear market. A Federal Reserve survey of bank lending practices showed credit tightening very sharply and broadly. Finally, the Institute for Supply Management (ISM) index for the service sector collapsed in January.
The spate of dismal reports touched off near panic in Washington. The Fed cut the “fed funds” rate by 1-1/4 points in late January, and Congress sped a $168 billion stimulus package through the often-lengthy legislative approval process.
Many forecasters in recent weeks revised their forecasts to now show a mild recession lasting through the first half of this year. They point to timely monetary and fiscal stimulus as the reason for a short and mild recession. They note though the possibility of a double-dip recession early next year as the stimulus fades.
Arizona’s economy is one of the most cyclical in the nation as measured by the difference between growth during expansions and recessions. The difference has less to due with losses during recessionary times than to explosive growth during good times. The volatility is due to swings in residential and nonresidential investment, which have an “accelerator effect” on the economy.
This is best understood by realizing that in a growing economy, resources need to be devoted to building infrastructure (new houses, offices, grocery stores, roads, etc.). When the economy slows, fewer resources (construction workers, cement, steel, brokers, mortgage lenders, etc.) are needed. That is, the resources that are tied to growth need to shrink! And if the economy stops growing, there is no need for these resources at all. But as this segment of the economy shrinks, there is less demand throughout the economy, and that means slower growth, which means less needed resources, which means slower growth, etc. In short, the investment component has an accelerator effect on the overall economy.
Since Arizona has such rapid underlying economic growth, a large portion of its economy is devoted this investment function, and that makes Arizona’s economy one of the most volatile in the nation.
The current business cycle was first and foremost caused by an inventory cycle in residential housing. The credit cycle came later as a result of the housing inventory problem. High energy prices also are contributing a third shock effect.
With this in mind, it’s clear that Arizona, along with a handful of other states (California, Nevada, and Florida) that have a large proportion of their economies related to growth, is leading during this cycle downturn. The question is whether it will lead at the trough as well. So far there is no indication that the investment cycle has reached bottom.
Inventories of resale homes listed on the MLS, as measured by months supply, are still increasing – nearly 12 months in Tucson and almost 15 months in metro Phoenix. Until supplies fall toward the “normal” four to five months, homebuilding will remain depressed. Housing prices will need to fall back to where they were at the end of 2004 to return to “trend.” That could be attained by the end of this year given the recent free fall of prices.
Current data for Arizona is consistent with a contracting economy. Seasonally adjusted employment, benchmarked to second quarter UI data, peaked in August of 2007 and through the end of the year lost 7,500 jobs, a decline of 0.3%. Employment is now declining in construction, manufacturing, information, employment services, financial activities, and transportation and warehousing. Official revisions will be released in March and are expected to be revised significantly downward (and close to our own benchmarking efforts).
Unemployment insurance claims continue to surge, and Arizona’s unemployment rate jumped by 1.2 percentage points in the final two months of the year. That puts Arizona’s unemployment rate at 4.7% -- only three-tenths lower than nationwide.
Retail sales in nominal terms peaked in January 2007. After declining sharply last spring, sales leveled off in recent months. With data through November, sales have fallen 2.5% during the past 12 months.
The Institute for Supply Management’s index for Arizona plunged as 2007 came to an end. Both production and orders components are now below 50. The last time that happened was during the 2001 recession.
Business bankruptcies (as well as personal filings) are now moving higher after tightening of bankruptcy laws a couple years ago and strong economic growth drove filings to decade-low levels. Chapter 11 reorganizations are now back at levels last seen in 2003.
Arizona consumer confidence, published by Behavior Research Center, plunged in their recent survey to mirror the large declines in nationwide confidence. The Arizona measure is now collected only twice per year rather than quarterly as in prior years. At a reading of 80, confidence now stands at the lowest readings since the end of 2003, during the “jobless recovery.”
Nonfarm employment is perhaps the best single indicator for state and local economies. How did this measure perform during past recessions (Exhibit 2)? During the 1960-61 and 1970 recessions, employment increased by 2.2% and 2.7%, respectively, reflecting the fact that Arizona’s economy was at that time insulated from the effects of national recession. That’s no longer the case as Arizona’s economy has grown and diversified significantly. During the 1974-75 recession, nonfarm jobs declined 2.1%, the largest ever recorded. Another large decline occurred during the 1981-82 recession when a loss of 1.8% was experienced. The 1990-91 recession saw jobs increase once again, by 0.5%. Then a decline of 1.2% occurred during the last recession of 2001.
In the recessions where jobs declined, Arizona lagged behind each cycle peak, i.e., the peak in Arizona came later than nationwide – by an average of four months. Timing at the low point was much closer, averaging less than a one month lag. This means that periods of contraction in Arizona have been shorter -- by about three months.
The U.S. Department of Labor recently began compiling data for states that shows dynamic labor market flows that underlie net changes in aggregate employment levels. These new data track the number of private sector jobs gained from opening and expanding establishments and the number lost at contracting and closing establishments.
Normally, job gains exceed job losses (except during recessions). For example, during the rapid growth experienced in 2005, Arizona’s net change averaged nearly 34,000 jobs per quarter (175,000 jobs added, 141,000 lost). During the last recession in 2001, job losses exceeded job gains with over 13,000 jobs lost (net) on average per quarter. For the past three quarters (4Q2006 – 2Q2007) job gains and losses have been nearly equal (Exhibit 3). 
The second quarter of 2007 is the most recent period for which the business dynamics data is available. It is based on unemployment insurance filings, which have a reporting lag of six months or more.
The data also show that roughly 3.5 times the number of new jobs is created in existing firms rather than in new firm openings. For example, on average in 2005, existing firms added 134,000 versus 40,000 in new firms per quarter. A similar relationship also exists for firms that are shrinking their workforce. Over the past year-and-a-half the number of jobs being trimmed in existing firms has moved significantly higher (Exhibit 4). 
The current contraction in the nation’s economy is likely to stretch past mid year and qualify as the 8th recession since 1960. Due to the monetary and fiscal stimulus being injected into the system, this recession should last eight months, give or take a couple, and be milder than average. Global Insight, a world-wide leader in economic forecasting, is forecasting a decline in real GDP comparable to the 2001 recession at less than 0.4%.
The recession in Arizona also should be mild with job losses totaling 1.3%. However, we expect that it will last longer than average, given the nature of this downturn and the length of time expected to get residential and nonresidential investment rolling again. We look for the bottom in the second quarter of 2009.
For a period of nearly three years during and following the 2001 recession, it seemed as though the bad times would never end (remember the “jobless” recovery?). Look for a repeat of that experience. 2008 will be remembered as the year you want to forget, and it will be late 2009 before the good times start rolling once again. |
|