Posted by: Josh Lehner | March 31, 2015

Strong Job Growth and Forecaster Bias

The good economic news continues to roll in. “[Oregon is] getting close to full-throttle economic growth,” as I recently told the Seattle Times. The reasons are the strong job gains and accompanying decline in the unemployment rate in the past few months. Furthermore, wage growth is stronger locally as well and migration trends have returned. As these economic gains continue to spread out across the state and across industries, Oregon’s recovery has turned into a nearly full-blown expansion. Ahead of our forecast advisory meetings, I just wanted to highlight a few thoughts on the recent numbers and one big issue our office is wrestling with.

First, these graphs help put the recent job numbers in comparison to past cycles. In absolute terms, 55,000 jobs over the past year is just about as strong has Oregon has ever added jobs in expansion. It has taken a few years of steady acceleration to reach this point, however job growth today is quite strong. Of course in percentage terms, growth rates are lower today because the economy and overall population is larger. However, given demographic trends, job growth of just over 3 percent is not only on par with the housing boom, but effectively full-throttle growth.


This is all good news for the Oregon economy. However, the data flow — to the extent that it’s accurate — is better than our office’s forecast. This is important, not just for selfish reasons that we want to be right, but because the upcoming forecast sets the revenue number for the 2015-17 biennium (and the kicker.) We need to get a handle on the underlying economy, so we can produce the best forecast possible.

There is also a fine line for forecasters between “sticking to your guns” and “rooting for your forecast.” Our office needs to make sure we come down on the right side of that line. In recent years, our office has not fundamentally altered the outlook and luckily we has also largely been accurate. As we approach the May forecast, with job numbers better than expected, what do we do with the outlook? It may or may not be too simplistic to simply answer “raise the forecast.” What if the stronger job growth is just a blip — notice the up and down cycle in growth rates even at the top of the housing boom — or worse yet, what if they are revised away? At the industry level, when does the manufacturing cycle end? These are the questions we will try to answer in our advisory meetings and have big implications.


Which brings me to the last point. There usually are a few funky aspects to the job reports in real time. This is certainly not the Employment Department’s fault, and it is not even BLS’ fault really (they’re the ones doing the survey and crunching the numbers.) It is just the nature of survey data and trying to weight it properly to reflect the whole economy. Doing this on a monthly basis, in real time is challenging. That’s why our office relies much more on the larger but less timely QCEW data, along with the monthly withholdings out of Oregonian paychecks, and we wait for the revisions to the monthly employment data. Case in point being the goofy seasonal factors last year — see June and December in particular. After revisions, those data make a lot more sense today. However one can identify these issues in real time. Our office has not put the latest job numbers through its paces just yet, but can spot a potential issue with the latest unemployment rate figures.JobGrowthURDropFeb15Over the past two months, Oregon’s unemployment rate has dropped 0.9 percentage points. Never before in Oregon’s recorded history has the unemployment rate declined by this large of a degree. Also, the latest, annual revisions to this data series went through Dec ’14. Most likely it is not a coincidence that the outlier data points also happen to be the ones that have not been benchmarked or revised. This, like the seasonal factors before, speak to underlying data issues, not the underlying economy.

To bring this full circle, it is possible that Oregon did experience record job growth a couple months ago and now record declines in the unemployment rate. That is a possibility and has a non-zero probability. However, it is also possible that the underlying data is very good, as it has been in the past year or so, but maybe not quite record good. It serves no one well for our office to not change the forecast because we are rooting for our current outlook. And, we are certainly not opposed to raising the forecast — again, our job is to be right, whatever the outlook — but we want to make sure it is for sustainable, economic reasons. Not purely because the last couple of months of data is above forecast. To be sure, there are clear signs for both near-term optimism and longer-term pessimism for the economy. I will post some thoughts on these issues as our advisory meetings get closer.


  1. […] follows is a mostly graphical update on the landscape of jobs across Oregon. As mentioned in the previous post on the state more broadly, Oregon is near full-throttle rates of growth. This acceleration has […]

  2. […] discussed back in March, Oregon job growth has been above forecast for the past 6 months or so but our office had been […]

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s


%d bloggers like this: