Economist Bill Watkins Notes Corrections to Forecast Article

I’m writing to correct several errors in an article Cascade Business News published electronically recently.

It is this article with the headline California Economist Negative on Oregon’s Future:

http://www.cascadebusnews.com/news-pages/e-headlines/2980-california-economist-negative-on-oregons-future

The first error occurs in the first two sentences of the third paragraph: “On the CLU CERF website the forecast stated that recent data shows that Oregon lost 7,900 jobs in September, and almost 6 percent annual job loss rate. CBN was not able to confirm where this number came from.”

We got the number from Oregon Employment Department’s website. The number is top center of this page: http://www.qualityinfo.org/olmisj/CES?areacode=41010000001&action=rs54&submit=Continu.

and beyond. I just did a Goggle News search and found many articles, including this one by Bloomberg Business Week News http://www.businessweek.com/ap/2012-10-16/report-shows-oregon-losing-jobs-in-september and this one by the Oregonian http://www.oregonlive.com/money/index.ssf/2012/10/oregons_growth_streak_snaps_in_september_economy_cuts_7900_jobs.html.

, italics in original. I was, of course, talking about the DX1 project which is widely reported to be $3 billion. I don’t know where the $10 billion number came from. Perhaps you added the expansion announced Wednesday October 24, days after my piece was written. In any event, here are two articles documenting the $3 billion DX1 facility and the newly announced expansion: http://www.bizjournals.com/sanjose/blog/2012/10/intel-plans-to-double-3b-oregon-rd.html & http://www.oregonlive.com/opinion/index.ssf/2012/10/intels_second_dose_of_expansio.html

The next sentence of your article continues “In actuality Central Oregon’s seasonally adjusted unemployment rates dropped slightly for all three Central Oregon counties in September.” In context, that is a non sequitur. You go from an Oregon jobs number to a Central Oregon unemployment number. You then extend the non sequitur by switching to unemployment numbers for Oregon and the United States.

I spent a fair amount of time trying to understand what you were trying to say here. You may have been referring, obliquely, to our use of non-seasonally-adjusted data for Central Oregon unemployment numbers.

All data are estimates, and when data are not particularly seasonal, the seasonal adjustment adds volatility and errors. When you look at Central Oregon unemployment data, they do not exhibit a strong seasonality. We therefore believe that the non-seasonally-adjusted numbers are more likely to be closer to the truth. Below, I’ve put a chart of Deschutes County unemployment data. Jefferson County’s and Crook County’s data are similar. There is no strong seasonality in the data: sounds more quaint and rural),” italics in original. Server farm is a commonly used term. Here’s an example from The Oregonian: http://www.oregonlive.com/silicon-forest/index.ssf/2012/10/t5_data_centers_plans_hillsbor.html. I also note that my use of the term included the planned T5 facility in Your sixth paragraph in the Real Estate section reads:

“There are other signs that Oregon’s residential markets will remain soft. Delinquencies and foreclosures remain high. Given the delays in Oregon’s unusual foreclosure process, it will take a while to work through the overhang.” Watkins charts on his website show foreclosures declining. Italics in original.

I guess the point of the italicized sentence is to imply that declining foreclosures are somehow inconsistent with soft markets or still-high delinquencies and foreclosures. There is no inconsistency here.

Then, there is your penultimate paragraph:

“In the meantime, we expect market conditions to improve only slowly in most of Oregon’s commercial markets.” Watkins makes this remark and reverse: “Oregon’s commercial building activity has climbed through the roof. The values reached $5.8 billion dollars in 2011 Q4, dramatically exceeding the value of $3.6 billion in the boom years of 2007. We estimate that by the third quarter of 2012 it will reach $6 billion.”

There are a couple of problems here. In the first place, I didn’t write the portion that comes after reverse. That was written by my colleague Dan Hamilton, and it is sourced at the top of the page. There is a larger problem. You are implying some inconsistency. There is no inconsistency.

Here are your two prior paragraphs:

He says Oregon’s commercial market conditions are truly unusual. While demand is mostly weak, the state is seeing huge amounts of new construction. The construction, though, is only in specific purposes. Intel’s Hillsboro project is a large component of the value of new commercial construction. The other source is data centers.

Watkins believes that while Intel and server farm facilities will create some local demand, they will not generate enough local demand to drive Oregon’s commercial markets to a robust recovery. “That will have to wait until Oregon sees a more broad-based jobs recovery. This is something that we don’t see happening real soon.”

I don’t think the analysis is so nuanced as to be difficult. Oregon is seeing lots of construction value in Intel projects and server farms. There is not much strength in other commercial markets, in construction, lease rates, or absorption.

Would you please run this letter as a correction, at least as visible as your original piece?

California Lutheran University Center for Economic Research and Forecasting Executive Director Bill Watkins