We peg the beginning of the Alaska recession at October of 2015. That is the point we identify as when the general economy began a downward trajectory. Prior to October 2015, we can identify some leading indicators that the economy was in trouble. But the data doesn’t show the economy as declining until that point.

We also think the recession is near its end, or maybe already over. Preliminary data shows signs of growth in 2018, even as the ripple effects of the 2015 disruption finish their path through the economy.

Employment Data

Let’s look at the data:

There are two data sources charted on this graph. The red line is the total number of jobs reported to the Alaska Department of Labor and Workforce Development (DOLWD) by employers that are subject to the unemployment insurance statutes.

The blue line is the total employment reported by the Federal Bureau of Economic Analysis (BEA), which gets its data from the Federal Bureau of Labor Statistics (BLS). These data include survey responses the BLS collects to estimate self-employment and other non-reported employment.

As you can see, both datasets tell a similar story. Until 2015, the job count in Alaska had been growing over time, with the exception of 2009 (when jobs were lost around the globe).

In 2016, Alaska experienced job losses of about 5,300 jobs. While the BLS data isn’t publicly available for 2017, data from the State suggests that another 5,200 jobs were lost. Preliminary data for 2018 suggests that job losses are still occurring, but at a slower rate.

Here is the 12-month moving average (to smooth out seasonal factors):

From the high point in September 2015 to the low point in May 2018, we estimate a total net loss of 11,483 jobs (3.4% of all employees).

Job Losses by Major Sector

The Department of Labor reports jobs in 11 major sectors. This graphic shows how employee counts changed in each of those major sectors during the last 3 years.

Notice that the State economy as a whole did not lose jobs in 2015 (blue bars). However, some industries did start to show signs of weakness. Looking at monthly data, it appears that losses actually started in the second half of 2015, but don’t look too bad as an annual number due to gains earlier in the year.

In 2016 (red bars), the major damage was inflicted on the mining (mostly in the oil and gas subsector), construction, and professional services sectors. 

In 2017 (yellow bars), the sectors most impacted in 2016 continued to lose jobs, but at a slower rate. You also start to see the secondary effects of those primary job losses creep into other sectors. Notable are the government and trade sectors.

It’s also worth pointing out that throughout the recession, the tourism and healthcare industries continued to grow.

Job Losses by Region

Of the 6 economic regions that DOLWD tracks, 2 experienced more significant job losses during the recession than the other 4 regions. The drop in employment is very noticeable in the Southcentral (Anchorage) and the Northern (North Slope) regions. This is likely because these two regions are more closely tied to the oil and gas industry.

Here is a plot of the 12-month average job count in each of the economic regions (Anchorage is shown on its own graph to avoid axis distortion):

For a better look, here are some graphics showing total job changes between now and when the recession started. The doughnut graph is total job losses. The bar graph is in percentage terms to normalize the data:

Wage Data

Some jobs pay over $100k per year. Other jobs pay the $9.84 minimum wage. Looking at just the employment data, these jobs are counted them the same.

While every job is important to the individual that holds it, some jobs are more important to the economy than others. The reason is that a person’s wages are what allows them to make purchases. Those purchases require labor on the other side of the transaction.

So, when the economy loses a $100k job, it means there are a tens of thousands of dollars that stop being spent on groceries, restaurants, bars, movies, clothing, etc. If less of these things are purchased, businesses need less people to sell them. This implies that losing high paying jobs will result in also losing lower paying ones.

Our analysis suggests that for every $100k oil job that is cut, the rest of the economy experiences $61k of lost wages about a year later.

It is clear on the graph that Alaska entered a recession in 2015. It is also clear that the recession appears to have ended. Beginning in the second half of last year, losses in wages started to turn around. The preliminary wage data from the first quarter of 2018 looks like it is continuing to rise. While employee counts haven’t started showing this trend quite yet, we think it is on the way.

From the high point in Q4 2015 to the low point in Q2 2017, the economy lost a total of $1.3 billion in annual average wages (6.1%).

Wage Losses by Sector

In 2015 (blue bars), only the oil sector showed signs that a recession was beginning. Some weakness was apparent in the professional services and construction sectors, but not enough to ring any alarms. The economy as a whole added wages in 2015.

Over $1 billion in total wages were lost in 2016 (red bars). While almost every sector felt the impacts of the recession, wage losses are largely concentrated in the non-renewable resource (oil), professional services, and construction sectors. Growth in the healthcare and tourism sectors offset a significant portion of those losses, and blunted the impact on the rest of the economy. (Note: these categories are different from the employment data because the come from a different source, but you can get the idea of what matches up to what).

2017 (yellow bars) saw continued wage losses in the oil industry, professional services, and construction sectors. However, those losses were much smaller than the year before. All other sectors showed wage growth during the year.

So far in 2018, things look to be improving. Here is a stacked graph of the net economy since the start of the recession. I know that’s a hard graph to read (you can download the data if you like).

The story this graph is telling is that pretty much every sector showed net wage growth over the last three quarters. If we are correct (and another economic shock doesn’t occur before then), these increases will start feeding additional increases over the next year. 

Conclusions

1. We peg the beginning of the recent Alaskan recession at the beginning of the fourth quarter in 2015. That is about a year after oil prices started to collapse.
2. The recession was almost completely constrained to the oil industry and it’s contractors. Secondary effects were more evident in job counts than total wages.
3. The recession impacted the Northern region more severely than anywhere else (due to its reliance on the oil industry for jobs).
4. Anchorage experienced a significant disruption in employment and wages across many sectors. But it appears those numbers are beginning to recover.
5. The rest of Alaska only felt “aftershocks” of the recession.
6. The government sector did not contribute to the recession by as much as commonly believed. While the State government did make cuts, they were minimized by drawings from savings accounts. They were also offset by increases in Local government employment.
7. Alaska’s economy is more robust than commonly believed. The general economy weathered a significant reduction in employment by the oil industry surprisingly well.
8. We believe that the recession ended toward the beginning of 2018. Wage and GDP growth are now signaling increases, and employment numbers appear to be stabilizing. As oil prices have greatly recovered, we believe an expansion in the oil industry is in its early stages. That expansion should feed into an expansion of the rest of the economy.
9. While the first half of 2018 will likely show continued job losses, we believe they will be minor compared to 2016. We expect the second half of 2018 to show modest growth.