Over the past year our office has developed a demographic alternative scenario to assess the state’s economy and public revenues should migration not rebound as expected in the years ahead. We included a preliminary assessment of the economic impacts back in our May 2023 forecast document. We continued to work on it and to incorporate the expected impacts on state revenues. This full-fledged scenario was included in our December 2023 forecast. We also discussed this with the Legislature at our forecast release. You can watch that discussion here (starts at the 52:55 mark and goes for about 30 minutes). The Executive Summary is copied below, with the full report linked at the bottom.
Population growth is the main reason Oregon’s economy outperforms the typical state over time. The influx of young, working-age residents allows local businesses to hire and expand at a faster rate. During the pandemic, population growth in Oregon slowed as it typically does during recessions, but so far in expansion it has not picked up. In fact, the Census Bureau estimates Oregon lost population in 2022 and 2023. These are the first estimated losses in nearly 40 years. And would be only the fourth and fifth losses since World War II.
Our office’s baseline forecast calls for a modest rebound in migration and population growth in the years ahead. Oregon is still an attractive, scenic place to live and there are plentiful job opportunities. However, Oregon must be open to the possibility that even a modest rebound in migration may not materialize in the years ahead. The pandemic may be a structural break point from an historical perspective, even if it will take a few years to fully realize it. Over the past year our office has developed a Zero Migration, demographic alternative scenario to scope out the potential changes the lack of migration will have on the state’s population, economy, and public revenues. For now it remains just that, an alternative scenario. It will take time to gather the complete data, and to assess future expectations.
In a Zero Migration scenario, Oregon will not change overnight. Each year, the relatively small differences between no net migration, and the modest rebound in the baseline may be hard to see. However, these seemingly small differences do accumulate over time. A decade from today, the state’s population would be outright smaller than it is today. The labor force likely would be smaller as well. Local businesses will have significant difficulty finding workers. That said, with increased participation rates among existing Oregonians, particularly those that help reduce historical disparities, the labor force could continue to increase even as the working-age population declines.
Even so, the state’s economy will not crater. Rather, Oregon would be on a slower growth trajectory. In terms of personal income, consumer spending, and public revenues, those will not decline outright. This is due to ongoing wage and income gains for existing residents, in part due to inflation, and also due to rising asset values over time. That said, the cumulative difference between the baseline outlook and the Zero Migration scenario does add up. Over the decade ahead, state revenues are expected to be $8.6 billion (-4%) below the baseline.
Such an impact is certainly large, and would require future budgetary adjustments. However, it is not insurmountable either when considering a smaller population will also have less demand on certain public services. One reason why the Zero Migration scenario impacts may be less severe than our office initially feared is that the baseline outlook is already weak from an historical perspective. Removing modest population gains has less of an impact than if the baseline called for strong growth. A second reason is the rising, nominal incomes and taxes paid. A third reason is timing. Our office produces a 10 year forecast. And in the first decade, the largest changes would be among younger populations. Oregonians in the late 40s through early 60s have the highest incomes and tax bills, suggesting the Zero Migration impacts would be larger in the second and third decades than in the first.
As a new year begins, our office is taking a three-part look at the economic outlook. Last week was a look at the U.S. economy. Today we dig into the Oregon economy. In two weeks we will look at the Oregon housing market.
In recent years our office has talked a lot about labor and capital being the key drivers of economic growth. So far this cycle Oregon has experienced good overall economic and income gains. However the composition of the growth is a little bit different in that Oregon’s employment growth is in the middle of the pack across all states, while our productivity has been stronger than most states. The combined result is that overall good growth. Looking forward it will be important to see how these factors change and how Oregon differs from other states.
Let’s start first with capital. There are five main types of capital — financial, human, natural, physical, and social — all of which can and do make workers more productive. In terms of something like GDP we tend to focus primarily on business investment in physical capital (plants, equipment, software, etc), although we know all sources of capital are important and a more skilled workforce provides broader economic gains and the like.
As of a month ago, Oregon’s productivity gains had been among the strongest in the nation. However recent data revisions have brought Oregon closer to the pack. In particular, the BEA did their comprehensive historical revisions, along with a few tweaks to the methodology, and when those filtered down to the state data, Oregon’s GDP did see a sizable downward revision to the previous estimates. If we look at 2022 and early 2023, only Wisconsin saw a larger negative revision than Oregon did. These downward GDP revisions for Oregon were entirely about manufacturing and real estate.
That said, something like real GDP per worker still shows Oregon having strong gains, just not as strong as previously estimated. As we highlighted in our most recent forecast, in the previous estimates with data through 2023q1, Oregon’s real GDP per worker was the 5th strongest nationwide. In the revised data, including using the latest employment data as well, Oregon’s gains are 17th fastest across the nation. From 2019q4 through 2023q3 Oregon’s real GDP per worker increased 5.3% overall, compared with the median state at 3.1%, and the U.S. overall at 4.9%. Encouragingly, Oregon remains above the pre-pandemic trend.
Note that while real GDP per worker is a crude measure of productivity, better metrics like the BLS’ labor productivity by state is available only once per year. As of 2022, Oregon’s productivity growth ranked 3rd best among all states over the cycle. The 2023 state data will not be available until May. And it must be noted that the BEA GDP estimates for goods-producing industries are used in the BLS measures (which are then also adjusted for hours worked, etc) so it is possible Oregon’s relatively rankings will be lower as well, but we shall see.
All of that said, ongoing business investment alongside increased skills of the workforce are needed to grow Oregon’s economy given the outlook for labor. Just in terms of the mechanics of growth, net job gains are likely to be slow to minimal in the year(s) ahead. The reason is twofold. First, demographics are mediocre at best. The large Baby Boomer generation continues to retire and replacing them in the labor force is a combination of Gen Z (smaller in Oregon than the Millennials), and the slower population outlook. Second, employment rates are high and unemployment rates are low. Nearly all Oregonians who want a job, have a job, or can more readily find one with the strong economy and tight labor market. Cyclically, there is no slack in the economy meaning businesses typically have a small pool of candidates when looking to hire.
This next chart takes the monthly employment numbers from the Oregon Employment Department, and compares them to our office’s estimate of the potential labor force. You can see that Oregon is near a record high by this measure today. This is a clear indicator of the tight labor market, as are other items like age-adjusted employment rates and so on. It is harder to find workers simply due to the fact that most everyone you would expect to have a job, actually has a job.
Given this, our forecast for employment growth moving forward is essentially tied to working-age demographics. For example, in our most recent forecast, our employment outlook from 2023q4 to 2024q4, so measuring the changes over the course of the year, was just 0.7%. That’s entirely based on demographics, and the modest rebound in population growth we have in the baseline (more on that in a minute). Also of note is that we had 2023q4 as the last quarter where job gains outpaced the underlying demographics a bit. Given the latest employment report for November which showed job losses in Oregon, it’s possible this process of slow gains tied to fundamentals once you reach full employment and you hit that labor constraint have kicked in a few months earlier than we had built into the forecast. We shall see.
So what are firms looking to hire supposed to do? They can increase their business investment to make their workers more productive, and they can also dig a bit deeper into the resume stack, hiring workers with an incomplete skill set and increase on-the-job training and the like. Of course they have to be able to find the applicants first.
That’s where Oregon’s Latent Labor Force comes in. We may be at the typical labor force constraint today, but that does not mean there are no more potential workers entirely. See our previous report for more details, but this untapped potential from reducing historical disparities remains, and can provide a boost to the labor outlook in the years ahead.
In updating Oregon’s latent labor force outlook, it finds that there remains a sizable pool of potential workers already living in Oregon when it comes to addressing and reducing disparities based on sex, race and ethnicity, and educational attainment. Now, a bit of good news here is that these updated estimates of the latent labor force are a bit smaller than the last time we did this. Partially this is due to a slower population outlook (not the good news), but also this is because these disparities are smaller than a few years ago, meaning some progress has been made, and the gaps still needed to be reduced are smaller.
In order for Oregon to see stronger employment and labor force gains this year, we need to see a stronger rebound in migration and/or increased employment rates on a demographically-adjusted basis. Such gains would likely be part of the latent labor force already living in the state.
If population growth does not rebound as expected, increased business investment and the latent labor force will take on an even greater importance for local businesses looking to expand. Given the new 2023 Census population estimates, and the work our office has been doing on a zero migration demographic alternative scenario over the past year, next week we will post that here on the site. It provides another potential look at the growth path(s) of the state, and is the reason why the housing outlook is coming in two weeks time.
Finally, an open question heading into the new year is to what extent we see income inequality decline. During the pandemic income inequality increased initially with booming capital gains and the loss of many, generally lower-wage jobs. (Remember that the recovery rebates were technically advanced federal tax refunds, which boost disposable income but not earned income which matters tremendously depending upon the data set you are looking at.)
Since then, we have continued to see the strongest wage growth among lower-paying industries. Here in Oregon, 17 of the 20 lowest-paying sectors have seen above-average wage gains since 2019. And nationally there has been a clear wage compression in the data. Wage inequality is smaller today.
However, as of the 2022 ACS data, this stronger wage growth has not translated into less income inequality at the household level. Income growth from 2019 to 2022 is relatively uniform across the income distribution, albeit lowest at the lowest incomes, and in Oregon at least higher at the highest incomes. At what point does the strong labor market, especially for lower wage workers, translate into improved household figures as well? Or to what extent do we see other factors, like increased household formation that lowers each newly formed household’s income, or the impact of retirements which drop more households down the distribution as we tend to earn less in retirement than in our working years and the like swamp the labor market effects? That’s what I’m watching for, and will continue to dig into, this cornucopia of socio-economic factors influencing the topline data we care about.
As a new year approaches, our office is taking a three-part look at the economic outlook. Today is a high level summary of the U.S. economy with particular attention on inflation and the Federal Reserve. Next week we will look at the Oregon economy, and two weeks after that the Oregon housing market.
Overall, 2023 was a good economic year. Employment rates are high, while unemployment remains lows. Importantly, underlying growth has been strong while inflation has slowed noticeably. This means income gains are now outpacing inflation which did not happen last year. Our office expects more of the same in the year ahead: high employment rates, modest employment gains that reflect underlying demographics, and then good income and wage growth that outpaces inflation.
In order to not bury the lede, the biggest expected shift in the macroeconomy next year is the Federal Reserve cutting interest rates. Right now the Fed is indicating they will cut rates three times, likely starting at the March or May meeting. However, if you look at the Taylor Rule — a rules-based formula to guide monetary policy based on how close inflation and unemployment (or GDP) is to target — we could be in for more rate cuts than that. Note that at this point there are many, many versions of the Taylor Rule based on different metrics and policy goals, see the Atlanta Fed for more.
If we take the Fed’s own forecast from their December 2023 Summary of Economic Projections (SEP), and plug it into a Taylor Rule (borrowing the one used by Renaissance Macro here), it points toward 4 rate cuts next year. And if you believe inflation is already at 2% (more on that in a minute) then the Taylor Rule suggests 6 rate cuts. Financial markets in recent weeks appear to have priced in something closer to this latter scenario, with low, but rising odds that the rate cuts may even begin in January. Keep in mind that these are not locked in promises but are really conditional forecasts. If the economy does X, then monetary policy should be Y. To the extent the economy differs from expectations, the Fed will too.
Given the past 18 months of high inflation and a Fed that was behind the curve and quickly raising interest rates to catch up, this marks a big shift in the U.S. economy next year. As we wrote in our most recent forecast, this shift was expected in general, but it could happen sooner than anticipated. Most forecasters, including the Fed itself were expecting a couple rate cuts in the second half of 2024. The key here is that the high interest rates were more about high inflation, so as inflation slows, the Fed can ease off the brakes while still maintaining relatively tight policy.
What has changed the outlook is the simple fact that inflation has slowed significantly more than expected. Inflation as measured by the West Region CPI has slowed to the 3-4% range both on a year-over-year, and month-over-month basis lately. That is still faster than pre-pandemic, but down sizably from 2022’s inflation numbers. And core PCE — the Personal Consumption Expenditures Price Index, excluding food and energy — which is the key measure the Fed watches and targets, has been near 2% for the past handful of months. Note that excluding food and energy is not because the Fed does not care about those prices, but rather due to their volatility they are not generally a good gauge of the underlying trend in inflation.
There has been a lot of talk about the last mile of inflation being the hardest, and right now that does not appear to be the case. It is an open question whether inflation truly is at target already, or if some of the slowing is transitory in nature with the true underlying trend still above the Fed’s target. But nevertheless, inflation has slowed a lot, which will allow the Fed to cut earlier, and deeper than many believed just a handful of months ago. Again, these expected rate cuts are simply following inflation down. Real rates, or inflation-adjusted rates will remain relatively tight, which depending upon your view could still count as “higher for longer.”
Now, the Fed does have a dual mandate of maximum employment and price stability. With inflation heading lower, both sides of the mandate are back in play in terms of impacting monetary policy decisions. Importantly the labor market has both remained strong (high employment and low unemployment) and has rebalanced. Measures like job openings and worker quits rates have returned closer to their pre-pandemic patterns which indicates the labor market is no longer overheated. This should continue to ease wage pressures and bring wage growth closer to rates consistent with the Fed’s inflation target.
One last thing to note is that even if wage growth remains relatively high, productivity gains can offset that and keep inflation in check. Supply side improvements this past year in terms of labor force growth and productivity gains have been able to bring inflation down without a recession, or period of low growth. That is why forecasters are now more optimistic about the outlook and recession probabilities than a year ago. This initial descent from the inflationary economic boom to the soft landing has gone as well as the Fed could have hoped for. The 2024 baseline outlook calls for the full descent and economic soft landing.
This morning Census released the 2023 state population estimates. The Census site has been down and I have been unable to access the data. However the Tennessee State Data Center has the topline state estimates on their site here. Using that information, we see that similar to last year, Census shows that Oregon’s population declined. Census estimates Oregon’s population fell 0.1% after a 0.4% decline last year. So while still a decline, a smaller one than last year. (While I trust the Tennessee State Data Center, and their Oregon numbers match what I can glean from Census’ social media, there is a small chance the numbers will differ once I can get the files directly off the Census site. I will repost should that be the case. UPDATE: As expected, the Tennessee data did match the Census data now that their site is up and running.)
In terms of the components, in 2023 Census estimates that Oregon’s natural population change was -4,200 as deaths outnumbered births, international in-migration was +4,300, and then net domestic migration was -6,100, for an overall annual change of -6,000.
In terms of annual growth rates, here is an updated chart out office uses regularly, along with our forecast. Our office’s baseline forecast is for a moderate rebound in growth in the years ahead.
Unfortunately, demographic data comes out infrequently and with a lag. As discussed in recent posts, we are just now analyzing the socio-economic patterns of migration in 2022. It will be another 10 months before we have any data on who moved in 2023, and then another couple of months to get the underlying microdata and analyze it and the like.
Given the noticeable differences between the estimates in recent years, it does give further credence to the possibility that Oregon’s population growth will not rebound even as slowly as it does in the baseline forecast. Our office has spent a lot of time over the past year thinking through and modeling the potential economic and revenue impacts if migration does not return. We spent time discussing this with the Legislature at the most recent forecast release as well. I will pull out that work and post it here in the coming weeks. None of this means we will suddenly shift our baseline forecast based on two years of different estimates. But it does mean we have to take that possibility more seriously, and to highlight the potential impacts.
Lastly, the Census site has been down today, but here is a state map of population change from Census to show the relative patterns across the country.
Today let’s take a quick look at maps that zoom in from the regional, to the state, to the local level based on the newest ACS data. These maps cover a range of topics that I have been discussing in presentations lately. For all of this work, I am using the tidycensus package in R, created by TCU geography professor Kyle Walker, along with a label editing assist from ChatGPT to help with the code. You can learn to use tidycensus here. For all things maps and data, Professor Walker is a great follow on social media (e.g. here is a recent map he created of drive times from Nike’s Beaverton headquarters).
Note that these first three charts will be using the 2022 ACS 1 year estimates, which are available at the Public Use Microdata Area (PUMA) level which are geographies of about 100,000 residents. The fourth chart uses census tract level data for the 2022 ACS 5 year estimates.
Map 1: Pacific Northwest Employment Rates
Historically, cities are where human progress has been made. The impact of lots of people living near each other, interacting, sharing ideas and driving innovation in totality is greater than the sum of its individual parts. And while it is true that large metros on the West Coast are lagging many of their smaller metro and rural neighbors this cycle, the urban-rural divide remains noticeable. The first map below shows the employment rates among 25-64 year olds across the Pacific Northwest.
It is clear that urban areas still have the strongest, most vibrant economies today, even if the relative growth patterns in recent years have shifted some. Both Portland and Seattle have high employment rates, as do other metros like Bend, Boise, Bozeman, Eugene, Medford, Spokane, and the like. You can see how employment rates within the urban/city portions of these counties have higher employment rates than the surrounding areas. This speaks to the continued productivity and importance of large, diversified economies even in a world of more remote work and a broader dispersion of population given the outflows of large cities during the pandemic. Urban areas still matter.
Map 2: Working from Home in Oregon
One of the long-lasting impacts from the pandemic has been the increased ability and acceptance of working from home (WFH). In 2022, 15% of the U.S. workforce worked from home most of the time, while it was 19% here in Oregon.
These impacts are seen in all of our communities. That said, one major shift in this data in recent years has been the relative pattern of WFH. The highest rates, and largest increases have primarily been within big cities. At first, the patterns looked like city residents sheltering in place. But now it’s not about the pandemic, but about changing workplace practices. Many workers still live in the big cities and suburbs, but are just no longer commuting into the office as often. As such, 13 of the 15 highest WFH share parts of the state are in the Portland Tri-County area with the Top 3 being in the City of Portland specifically (30%+ WFH rate). The other 2 in the top 15 that aren’t in the Portland region probably aren’t a surprise with the City of Bend (25%, ranking 9th highest) and the rest of Central Oregon (19%, 15th highest).
Map 3: Child Poverty in Portland
This next map shows the 2022 child poverty rate in the Portland region. It consists of all the PUMAs in Clackamas, Multnomah, and Washington Counties in Oregon, in addition to Clark County in Washington.
The concentration of child poverty in a few specific areas of the Portland region stands out. East Portland in particular has the highest rate of child poverty in the state at 25%. Overall the U.S. child poverty rate is 16%, and Oregon as a whole is 14%. High rates of child poverty are also seen in the Portland area in Oregon City/Gladstone (19%), Beaverton-Cedar Mill (15%), and in West Vancouver in SW Washington (14%) . Economic research has shown that one of the key factors affecting economic mobility – the probability a child is born poor but grows up to be successful – is economic and racial segregation. Children growing up in concentrated poverty are at a noticeable disadvantage. So even as the recovery from the pandemic has been broad-based and inclusive, it does not mean these disparities are gone.
Map 4: Southern Oregon Household Income
The final map of the data looks at median household income by Census tract across a broadly defined Southern Oregon region that includes Coos, Curry, Douglas, Jackson, Josephine, Klamath, and Lake counties. This data comes from the just-released 2022 ACS 5 year estimates. Right now the 5 year estimates are a little bit of a goofy time period as it includes a couple pre-pandemic years, the pandemic, and the initial recovery. But it is the most recently available data for places that are not part of large metros or counties.
In the 2022 ACS 5 year estimates, the U.S. median household income stands at $75,000 and Oregon is $77,000. Given that much of Southern Oregon is rural, you will see that most of these census tracts have incomes lower than the state or the nation. That said you do see higher incomes (darker colors) on the edges of Coos Bay, Roseburg, Medford, and Klamath Falls. Economically that’s part of the city impacts of the larger, more diverse economies showing up, and also I suspect the impact of higher-income households clustering in certain areas due to housing, be it neighborhoods with larger homes, or larger lots, or up on the side of the mountains with views and the like. But like a quilt, you will also see low-income census tracts patchworked right next to these tracts pointing out the wider range of socio-economic outcomes we have in our society.
In 2022, 77,400 Oregonians who were living in Clackamas, Multnomah, or Washington counties packed up and left the state. This represented 31,200 households.
In total, 9,300 Portland area household moved north to Washington (30% of all out-migrants), including 5,500 to Clark County specifically (18%). Another 5,900 households moved south to California (19%), with about a 60/40 split between northern and southern California destinations. After that, the major destinations for former Portland area households in order are Arizona, Texas, New York, Colorado, Utah, Florida, Idaho, and Illinois.
It is becoming increasingly clear that housing costs are a major factor in recent migration trends. The table below provides a breakdown of where former Portland area households moved to, including relative housing costs. These housing costs are estimates based on median home values in the PUMA (geographic areas of about 100,00 residents) in which they now live, compared to the median home value in which of the three Oregon counties they moved away from. (In the data you can get these households’ current housing costs — rents or home value — but you cannot get their former Portland costs, so we are using median values here to get the big picture relative cost differences.)
The upshot is that most Portlanders who left moved to somewhere with relatively lower housing costs. This general pattern is also seen in state level migration data, particularly among workers who are able to work from home leaving high cost of living locations.
Specifically, when looking at these former Portland area households, 15,500, or 50% of them moved to places like Arizona, Washington, Texas, and New York where their housing costs are now lower than in Portland. On average, the median home values based on these migration patterns suggest home prices are $200,000 or 36% lower in these households’ new locations.
Another 5,200, or 17% of the former Portland area households moved to locations with similar housing costs, or where home values are within 10% of Portland values. Most of these migrants moved across the river to Clark County, where median home prices are similar to Multnomah County, but generally lower than Clackamas and Washington Counties.
Conversely, 10,500, or 34% of the former Portland area households moved to locations with higher housing costs. Notably, most of these moves were south to California, or north to the Seattle Metro Area. Both of those markets are categorically more expensive than the Portland market. At first, this does appear to be a relatively higher share of migrants moving to more expensive markets than one may think. But in looking at 2018 and 2019 migration patterns, the relative share of households moving out of Portland to California and Washington is broadly consistent with the 2022 data. The difference is not the relative patterns per se, but rather the size or magnitude of the out-migration.
Finally, the last set of numbers in the table look at relative moving patterns and housing costs for former Portlanders moving to all other locations except for California or Washington. Admittedly this is a calculation that specifically excludes higher cost markets. But one reason this is a useful look at migration patterns is the so-called gravity model of migration, which suggests that people do not tend to move very far geographically. Among the 15,900 former Portland area households who moved away from the West Coast, there is an even starker pattern of moving to more affordable markets. 75% of these households moved to locations that, on a weighted median home value basis, are 40% less expensive than the Portland area. Only 14% of these households moved to more expensive locations, primarily the Denver, New York, and Urban Honolulu metro areas.
All of this is not to diminish the myriad reasons why people move. Moving is a very personal choice about where one believes the best opportunities for themselves, and their families may be. Households take into consideration things like social issues, or crime and homelessness, in addition to taxes and schools, among others. However, underlying all of these is the ability to financially make ends meet. Housing affordability is likely the biggest factor, although not the only one.
Last week our office highlighted the newly released 2023 population estimates from Portland State, and are awaiting the Census estimates in the weeks ahead. We are still combing through the 2022 ACS microdata to better understand what all happened from a socio-economic and migration perspective. Today we have a few things to note on migration trends across the state. On Thursday we will have some more specifics on Portland out-migration as well.
Note On Different Census Estimates
Our office regularly discusses the difference in estimates made by Portland State University and by the Census Bureau. However, it must be noted that Census itself reports different estimates as well. There is the commonly used annual midyear population estimate that only provides total counts of the population and the number of people moving. Then there is the more detailed American Community Survey (ACS) data which is collected throughout the year and combined into an annual figure. The ACS provides the socio-economic characteristics of the population and of migrants. The methodologies differ, as do the time periods (midyear point estimate vs annual average). That said, the ACS data is designed to be weighted to match the midyear estimates, but differences do arise.
The scatterplot below shows migration rates by counties across the U.S. using data from both the population estimates, and the ACS. As you can see, they broadly agree but there is clearly noise when looking at the specifics for any given county. In particular, a few Oregon counties are outliers. Annual population estimates showed both Deschutes and Linn gaining residents through migration, but the ACS showed they lost population. Conversely, Clackamas, Lane, and Marion all see stronger migration in the ACS data than in the midyear estimates.
Unfortunately, there is no one perfect number.
Migration Trends by Region
Using the ACS data, our office was able to create migration charts showing the inflows and outflows of migrants for different parts of the state. These are similar versions to the statewide chart our office regularly uses in presentations.
The bottom line is that while the trends are more pronounced in the Portland tri-county region (Clackamas, Multnomah, and Washington), they are more broad-based than some may realize.
Even so, the broad groups mask differences at the regional or local level. The scatterplot below looks at net migration among those moving within the state, and those moving between states. The most granular regional groups are so-called Public Use Microdata Areas (PUMAs) which represent the smallest geographies for which Census releases the underlying sample data. These are generally geographic areas of about 100,000 residents, which is why in some parts of Oregon some counties are grouped together.
In the chart, regions in yellow experience net out-migration, while blue regions saw net in-migration. The blue regions on the left side of the chart saw large enough within-state migration to offset losses of local residents moving to other states, while the blue regions on the right side saw positive net migration from other states, while within-state migration was relatively close to zero.
Keep in mind that the scatterplot uses ACS data which showed lower numbers for places like Deschutes and Linn counties, and higher numbers for counties like Clackamas and Marion.
Finally, given the discussion at the forecast release the other week, this last chart is a longer term look at migration in Southeastern Oregon. In the Census data, this region is the combination of Klamath, Lake, Harney, and Malheur counties. Overall, net migration across state lines is relatively constant for this region of the state, except for recent years where there is a net gain from other states (blue line above the yellow line). In fact, Southeastern Oregon, in the ACS data, is the only region in the state that saw both net in-migration among Oregonians, and net in-migration from those moving across state lines last year.
The preliminary 2023 population estimates for Oregon are out from our friends at Portland State University’s Population Research Center. These estimates will be certified in the weeks ahead, and may change some between now and then. Census has not set an official release date for their 2023 state estimates, but they are due out in December. Our office will post those when available.
But the big picture is as follows:
PSU revised down their 2022 population estimates for Oregon, going from a previously published +0.43% growth rate to a revised +0.09% growth rate. This is a downward revision of 14,000. In 2023, Oregon’s population is estimated to have increased 0.53%, or +22,600. It will be interesting to see where Census comes in for 2023 given the divergence, we are seeing in these estimates. And as we noted when talking with the Legislature the other week, we really won’t know where the population stands until the 2030 Census which is the huge, demographic anchoring point.
In terms of the county level estimates, the vast majority of the 2022 downward revision was in Clackamas County (-10,000) although 17 counties saw their estimates revised down in some form. 18 counties saw an upward revision, with Benton seeing the largest numerically at +2,600. Marion was essentially unchanged (-16, or -0.005%).
For 2023, PSU estimates 26 counties saw growth, while 10 counties experienced declines. Many of the slower increase or outright estimated declines are in rural counties. The stronger population gains are in urban counties. This is interesting to note given this is more of a return to pre-pandemic patterns, with the Portland region seeing stronger gains than the mid-Willamette Valley, and then Southern Oregon seeing a slowdown, and in Klamath as well which has outpaced the state in recent years. I suspect the declines in rural Oregon, like on the North Coast or in much of Eastern Oregon are due to demographics and an older, aging population and not so much about out-migration but we shall see when more details are released by Census eventually (county data from them will be out in March 2024). There are certainly exceptions of course, like the gains in Morrow and Umatilla, and ongoing stronger trends in Malheur.
Bottom Line: Our office’s baseline forecast calls for a modest return of population growth in the years ahead. These 2023 PSU estimates broadly fit that pattern, albeit after revisions, they are slightly below our forecast as it used the previous 2022 numbers as the base, but not enough to really alter the outlook. Given deaths outnumber births in Oregon, our state’s population would decline tomorrow without migration, it will be important to track these estimates from PSU and the upcoming Census release to gauge the underlying, demographic fundamentals of the state’s economy.
Finally, our office is still working through the 2022 ACS data and will have more on migration in the days ahead, following-up on some of our forecast release discussion with the Legislature.
On November 15th the Oregon Office of Economic Analysis released the latest quarterly economic and revenue forecast. For the full document, slides and forecast data please see our main website. Below is the forecast’s Executive Summary and a copy of our presentation slides.
** The forecast is corrected on November 20th due to overstating Insurance Taxes in the General Fund in 2023-25 by $117.7 million. All other revenues remain as published on 11/15. All forecast materials have been reissued to reflect the correction. **
The economy is rebalancing. Inflation remains above the Federal Reserve’s target. However inflationary pressures have eased as stronger productivity gains and in increase in the number of Americans looking for work means the labor market is no longer overheated. As wage growth slow, so too will overall inflation in the year ahead.
As a result, the Federal Reserve is now looking to make surgical rate cuts. As inflation continues toward target, the Fed can ease off the brakes, allowing the economy to coast. Most forecasters, including the Fed itself, expects a couple of small interest rate cuts in the second half of 2024, however the rebalancing labor market and cooler inflation readings may allow the Fed to cut sooner than expected. Lower rates will spur more activity, ensuring the economic expansion continues.
Oregon’s baseline economic outlook continues to call for the soft landing, and remains effectively unchanged from the prior forecast. Local economic growth is driven by a return to full employment, combined with stronger business investment and productivity gains. Recently released Census data confirm that the economic recovery from the pandemic has been inclusive and broad-based. Looking forward, a modest rebound in migration in the years ahead will allow local businesses to hire and expand at a faster pace than the nation. However, should migration not return as expected, Oregon’s economy will not crater, but rather grow at a slower rate than in the baseline.
Oregon’s state revenue outlook appears to have stabilized. Aside from persistently strong corporate income taxes, collections in recent months have tracked closely with the September forecast. In particular, personal income tax collections have finally started to weaken.
General Fund revenue collections are expected to decline significantly in the months ahead as corporate profits and business income return to trend, and a record personal income tax kicker credit is issued. Although the revenue outlook appears on track for now, Oregon has yet to go through its first personal income tax filing season of the biennium, so considerable uncertainty remains.
Even excluding the payment of the kicker credit, General Fund revenues were expected to be relatively unchanged when compared to the previous biennium. The revenue boom seen during tax year 2021 is unlikely to be repeated, with collections expected to revert back to their long-term trends. Traditional gains in General Fund collections are expected to resume in the 2025-27 biennium and beyond.
Longer term, revenue growth in Oregon and other states will face considerable downward pressure over the 10-year extended forecast horizon. As the baby boom population cohort works less and spends less, traditional state tax instruments such as personal income taxes and general sales taxes will become less effective, and revenue growth will fail to match the pace seen in the past.
See our full website for all the forecast details. Our presentation slides for the forecast release to the Legislature are below.
Short post this week that highlights the strong, cyclical labor market. The key point is that there is not a labor shortage due to fewer Oregonians (or Americans) looking for work, but rather due to businesses looking to hire even more, after a full recovery from COVID. Earlier this year we touched on the potential labor force and noted that yes, employment rates among teenagers and early 20-somethings is lower than it was back in 1999 or 2000, but mathematically the increased share of older adults working offset that in the big picture. Today I want to focus just the so-called prime working-age population, or those between the ages of 25 and 54 years old. Painting with a broad brush, this is a key demographic because it is when most people are finished with their formal schooling years, and before their retirement years. The vast majority of folks in this age range work, but the exact percentage depends on the strength in the economy.
This first chart shows the prime-age employment rate, or the employment-population ratio (EPOP for short) in Oregon for the past 40+ years. Notice how as of September, the only times in Oregon’s recorded history that the employment rate was higher than today was back in the 1990s. The same is true nationally. We are about 1 percentage point away from a record high, and provided the expansion continues, and given the mediocre demographics of a structurally tight labor market, I’d think we will reach that new high in the year or two ahead.
Note: Our friends at the Employment Department crunch the actual underlying data and share with us this measure from 2002 through current. Historical figures are based on our office’s work using the publicly available microdata.
Crucially, the recently released 2022 ACS data confirms the fact that the economic recovery from the pandemic has been inclusive. Based on this data, which begins in 2000 as the ACS replaced the old long-form decennial Census, Oregon’s prime-age EPOP is at a new high in recent decades, for both white, non-Hispanic Oregonians, and BIPOC Oregonians. This data is unable to compare to those previous all-time highs reached in the 1990s, but recent trends are certainly encouraging and point toward a strong labor market.
I still have more detailed work to do in going through the 2022 ACS data, but based on the data so far, there are still varied experiences among different racial and ethnic groups in Oregon. Employment rates for white, non-Hispanic and Asian Oregonians are higher than the overall prime-age EPOP. Hispanic Oregonians prime-age EPOP is right around the overall Oregon figure, while employment rates for both Black Oregonians and American Indian or Alaska Native Oregonians are below the statewide average.
Comparing across the cycle so far, both Black, and Hispanic or Latino Oregonians employment rates in 2022 are above their 2019 employment rates. However, the prime-age Black EPOP previous peaked in 2018, and Hispanic or Latino prime-age EPOP peaked in 2017. In 2022 their respectively employment rates were nearly all the way back, but just a hair under those previous values. On the other hand, employment rates for Asian Oregonians and American Indian or Alaska Native Oregonians continued to rise throughout the pandemic, and set new all-time highs, at least for data back to 2000.
This latest employment data by race and ethnicity in Oregon is similar to our previous look at the racial poverty gap, differences clearly remain, but they are a bit smaller today than prior to the pandemic.
Oregon Office of Economic Analysis 