Torto Wheaton Research Forecasting in a Rapidly Changing Economy; Base Case and Recession Scenarios June 2001

Transcription

1 Torto Wheaton Research Forecasting in a Rapidly Changing Economy; Base Case and Scenarios June The economy has shifted rapidly in with increasing risks of a possible recession evident. Torto Wheaton Research still believes that the US economy will not fall into a recession as classically defined. However to provide our clients with as much information as possible our TWR Office and Industrial Outlooks consider what the downside or a recession may look like. This paper reports the results under our base forecast and recession forecast to our clients. Further metropolitan details for each forecast can be found in the TWR Outlooks. The Outlooks ordinarily combine first quarter completion data with updated economic forecasts from March. This year economic conditions have changed so quickly that the Outlooks include not only the economic forecasts produced by Economy.com in March, but also a recession scenario. The recession scenario presumes that the final data for the year will show that the negative net absorption experienced in both the office and industrial markets in the first quarter was the result of a significant drop in employment. TWR is providing the recession scenario to clients not because we expect it to be the outcome by the end of, but to set a lower bound. TWR's base US scenario uses the prediction from Economy.com of 2.3 percent office employment growth in. This represents a significant slowdown from last year s growth of 3.7 percent. This employment growth estimate results in net absorption for the year totaling 60 million square feet, again a significant slowdown from last year s 116 million square feet for the 54 markets (72 MSAs) that comprise TWR's US market forecast. However, given negative net absorption in the first quarter, achievement of this outcome would require the pace for the second to fourth quarters to be equal to a little better than 25 million square feet per quarter. While this level is reachable in a growing economy as we saw in to, as we stretch into the second quarter with mixed economic news, such a pace seems less likely. For this reason, TWR has created an independent recession forecast, which is separate from the base case. This forecast begins with the assumption that the first half of the year will net only slightly better than zero for absorption. The employment forecast that would be required to achieve this result is then determined. As a result, the recession model uses negative 3 percent office employment growth for. Interestingly, this is approximately the same office employment decline seen in the recession year of, although the absolute numbers of employees affected is larger today. The growth rates of our base forecast are used in the recession scenario for 2002 and beyond. Figure 1 compares employment growth rates from the Office Outlook to the update made for Outlooks, and then to the assumption. The graph highlights how Torto Wheaton Research (617)

2 little Economy.com s forecast for office employment moved between December and March, as well as how extreme the employment forecast would have to be to give the first quarter net absorption numbers any sort of permanence. Figure 1: Office Employment Growth Rate (%) As such, the recession forecast is likely worse than what would occur if the economy improves in the last half of the year as is commonly believed. This belief is the key. The net absorption estimate under the recession scenario is reduced to 15 million square feet in, and it only improves to 30 million in As a result, the recession forecast shows vacancy rising to 13.6 percent by end of year 2002 with rents rising by only 1.9 percent, well below the rate of inflation. This contrasts to the base forecast, which shows vacancy rates increasing only slowly to 11 percent from the current 9.5 percent. Figures 2 and 3 summarize the three employment scenarios outlined earlier-spring, summer Base and summer effect on office real estate. With little change in the annual employment forecast, the baseline summer forecast differs little from the spring edition. As we know, the bears think otherwise! They correctly point out that the first quarter figures would require above average absorption in the final three-quarters to satisfy the original yearly forecast. Note also that the effect of the year of negative office employment growth translates into multiple years of slowed net absorption. Figure 3 indicates that this slowdown would takevacancy rates higher more quickly to a level just above equilibrium rates Torto Wheaton Research (617)

3 Figure 2: Office Net Absorption (x1000) 140, , ,000 80,000 60,000 40,000 20,000 - Figure 3: Office Vacancy Rate % As mentioned in our client release on the first quarter, negative net absorption could reverse quickly. It is commonly believed that the tech bust will end this quarter. Some reports had tech company bankruptcies peaking as early as February. Second, if other companies are indeed frozen by uncertainty, then the resolution of that uncertainty through interest rate cuts or a clear signal as to a bottom to the economic situation would result in net absorption proceeding forward. This logic would lead us to believe that the truth lies in between our two forecasts and suggests that clients observe both forecasts accordingly On a market by market basis, while our base forecast continues to use Economy.com estimates, the recession forecasts apply the negative 3 percent office employment growth rate indiscriminately to every office market. Differences in forecasts correspond largely to differing Torto Wheaton Research (617)

4 amounts of pent-up demand in each market, as well as the quickness of recovery in Economy.com's post- forecast. Markets hit hardest in the recession scenario include many of those with the largest recent run-up in rents; San Francisco, Ventura Co., Oakland, St. Louis, and Washington. In contrast, those with good underlying employment and population trends recover much more quickly and include Riverside Co., New York, Stamford, Phoenix, and Dallas. One final point. Even in the majority of recession forecasts rents are inflating. Clients should notice that TWR's historical rent series is not as volatile over the last couple of years as press reports may infer. TWR is reporting average, blended rents measured over a year in our TW Rent Index. This provides a different, but long term and correct, picture of the market of which we are writing. Many other accounts are using fourth quarter rates in contrast to current levels or in some cases are reporting marginal rents from recent leases as average rents. We suggest that in all the hype, clients should be careful to ascertain exactly what is being measured before looking at period to period changes. The Industrial Market Outlook For industrial space, the base scenario uses the prediction of 0.8 percent distribution employment growth, negative 0.5 percent manufacturing employment decline, and 3.1 percent growth in manufacturing output from Economy.com s forecasts. This represents a significant slowdown from last year s growth of 1.7 percent, 0.0 percent, and 5.4 percent respectively. These employment growth estimates result in net absorption for the year of 164 million square feet, again a slowdown from last year s 197 million square feet. However, given negative net absorption in the first quarter, this would require the pace for the second through fourth quarters to be higher than the quarterly pace seen last year. As we move into the end of the second quarter with mixed economic news, such a pace seems unlikely. For reasons similar to the office market, TWR has created a recession forecast presented separately from the base case. In this forecast TWR has started with the assumption that the first half of the year will net only 30 million square feet for absorption. The economic forecast required achieving this result was then determined. As a result, the recession model uses negative 4 percent manufacturing and distribution growth for while using negative 1.5 percent growth rate for manufacturing output. While this is slightly better than the recession year of for manufacturing employment and output, it is slightly worse than in the distribution sector. The growth rates of the base forecast are utilized for the recession forecast for 2002 and beyond. Figure 4 shows the growth rates for the distribution employment sector. Torto Wheaton Research (617)

5 Figure 4: Distribution Employment Growth (%) As in office space, a short recession has lasting effects. As such, the recession forecast is likely worse than what would actually occur if the production sector improves in the last half of the year as inventories decline. While the net absorption estimate is reduced to 94 million in, it remains at 67 million in As a result, the recession forecast shows rent growth in 2002 of 1.0 percent This contrasts to the base forecast, which shows availability rates nearly flat in with rent growth remaining above the projected inflation rate of 2.5 percent through Figures 5 and 6, highlights the slight difference to the rent forecasts with the few basis points shaved off between the December and March forecasts from Economy.com. The recession forecast is required to really make a difference to the outlook for rent growth. Because the industrial market sees construction adjust faster, rents revert to more normal growth rates visibly faster than the office market. Figure 5: Industrial Net Absorption (x1000) 250, , , ,000 50,000 - Torto Wheaton Research (617)

6 Figure 6: Industrial Rent Growth (%) Industrial space has is own expectations of reversal. As inventories are worked off and interest rate reductions start to take effect, many are hopeful that investment will resume in the second half of the year. This logic would lead us to believe that the truth again lies in between our two forecasts and suggest that clients observe both forecasts accordingly. On a market by market basis, while the base forecast continues to use Economy.com estimates, the recession forecasts apply the negative 4 percent employment growth rates indiscriminately to every industrial market along with the negative 1.5 percent output decrease. Differences in forecasts correspond largely to differing amounts of pent-up demand in each market, as well as the quickness of recovery in Economy.com post- forecast. Markets hit hardest in the recession scenario include many of those with the largest recent run-up in rents; Seattle, Indianapolis, Nashville, Austin, and Tampa. In contrast, those markets where increasing office rents have increased the value of the land component are little effected by the recession; including Oakland, Stamford, San Francisco, Orange Co. and San Jose. As TWR releases this report, we begin work on our fall forecasts for the TWR Fall Outlooks. Yet another round of Economy.com forecasts and the greatly anticipated real estate data from the second quarter will be examined to identify the impact on the econometrically derived relationships, which will in turn produce a new forecast. There have never been so many differing opinions on the economy as there are today. This uncertainty gives objective analysis real value in defining the parameters. As additional data provides more information on the resolution of the current slowdown, these parameters are sure to narrow. Torto Wheaton Research (617)