You’ve probably heard by now that Alaska currently has one of the highest unemployment rates in the nation. While this is true, it doesn’t quite mean what people think it does. Here a few reasons why.

1. Alaska has a higher “natural rate” of unemployment than other States

There are many reasons a person might not have a job. They could have gotten laid off. They could have just graduated and haven’t found one yet. Maybe they are retired, or in school, or in the hospital, or in prison. Or maybe they just don’t want to work right now.

For the purpose of public policy, we need to distinguish between those people who want to work from those that don’t. A person that spends the day on the golf course is not the same as one sleeping in the streets. If the economy has a job for every person that wants one, we should think things are going well.

Natural Unemployment

Actually, we think an economy is doing well even if there is not a job for every person. We understand that people sometimes take a few months between jobs. We get that someone may not want to work in a fast food chain when the lose their job as an executive. And we can forgive a business for not hiring someone without the skills to do the job.

Think about all the time you spent looking for work over your lifetime, Think about the time you took after college to find your career path. The time you spent switching jobs because your current job just didn’t fit your goals. Think about the time you were laid off because the economy took a bad turn. The time you spent looking for work in a new town when your spouse got a job offer. And what about the months you spent learning a new skill because your old skills were no longer relevant.

It adds up.

Natural Unemployment Rate

In your 40 or so years that you will call your career, there will probably be roughly 24 months in total that you were considered “unemployed.” We consider this normal. We don’t expect anyone to work for 40 years at one job (although some people do).

This acceptable time off is what economists call “natural unemployment.” It’s unemployment, but it’s not the kind that leaves you homeless. It usually ends up making things better off for you.

Since everyone in the economy is experiencing this same phenomenon, it tends to average out across the population. While you have a job, someone out there is trying to find a way to better use their skills. So, at any given time, if we ask everyone in the country if they are working or looking for work, we expect to find about 5-6% of the US population between jobs. This is what we call the “natural rate of unemployment.”

Alaska’s Natural Unemployment Rate

It turns out that Alaska tends to be consistently above that number. Over the last 40 years, the average unemployment rate in the US has been 6.3%. In Alaska, it has averaged 7.9% since 1976. In fact, in any given year you could count on Alaska’s unemployment rate being the National average plus 1.5%. Look at the data:

With the exception of the Great Recession (which hit the Lower 48 much harder than it hit Alaska), our unemployment rate is pretty much always higher than the rest of the country. Put another way, it appears that Alaska has a higher natural rate of unemployment.

2. Seasonality plays a role in pushing up the average unemployment

Alaska has a lot of jobs that only happen in the summer. This includes tourism, commercial fishing, firefighting, and others. These seasonal jobs create a great deal of fluctuation in Alaska’s monthly unemployment data.

For the US as a whole, the unemployment rate moves about 1.4 percentage points between its seasonal highs and lows. In Alaska, that range is 2.7 percentage points.

Why is this important? Because the official unemployment rate is a seasonally adjusted number.

When you have a job for half of a year, the unadjusted rate counts you as employed during six months and unemployed for the other six months. But the adjusted rate basically counts you as having half a job each month. Hence, the annual (and seasonally adjusted monthly) unemployment rate is higher on average.

Here’s another look at Alaska’s unemployment rates, divided between the summer months and the rest of the year.

So, part of the reason that Alaska has a higher natural unemployment rate is the seasonal nature of the economy.

3. Alaska’s smaller population makes unemployment rates less reliable

Like all statistics, macroeconomic indicators work best when there is a large population. This allows the anomalies and random factors to get diluted in the pool of data. In smaller datasets, the weight of each data point is exaggerated.

For example, a billionaire living in a town of 1,000 people would give the impression that they are all millionaires (on average). If you put that billionaire in a city of 10 million people, their wealth barely moves the needle of the city average.

Regional Factors

Take a look at Alaska’s unemployment rate separated between Anchorage and the rest of the state.

Clearly, the higher unemployment rate in Alaska is actually a higher unemployment rate outside of the metropolitan center. Anchorage unemployment rates tend to track the rest of the US quite well. There are a couple of reasons this might be. One of those reasons could be an issue with sampling.

Sampling Errors

Sampling is how we do statistics. For example, if we want to know if a soup is too salty, we can tell with a spoonful. We don’t have to eat the whole pot.

Say you have a town of a million people. You can sample 0.1% of the population and have a pretty good sense of what is going on. With just 1,000 phone calls you can be pretty comfortable that your results resemble reality.

But what if you have a population of just 10,000 people. Will calling 10 people tell you much of anything? And if you are trying to understand employment in a village of 100 people, you probably need to call everyone in that population before you can be comfortable with your numbers. And that’s if you get reliable data.

Say you make 10 phone calls to generate an inference about those other 90 people. On average, 8 will say they don’t have a job, and 2 will report working. If one of those 8 people say they are looking for work, you would have an estimated unemployment rate of 33% (1 unemployed, 2 employed). When you scale that number up to the rest of the population, you have a significant error.

How They Calculate Unemployment

This is why the BLS doesn’t bother trying to determine actual numbers in small populations. And that is why the numbers that are posted don’t mean as much as you might think.

Look at this statement from the Bureau of Labor Statistics on how they determine local area statistics:

“Both new entrants and reentrants are estimated from state-level CPS data on a 5-year weighted average for each month. Then, the averages for each state are controlled to the corresponding national total for the current month. Next, these adjusted statewide entrant totals are distributed to each county, using population shares based on the latest annual July-1 population estimates from the Census Bureau.”

In other words, it’s not an Alaska number at all, it’s a national average forced onto Alaska’s population. It’s also not a current month statistic, it’s based on a 5-year average. And it’s based on the Current Population Survey, a sample of 60,000 households across the nation hand selected to determine what is happening on a national level. It is unlikely that more than about 100 of those households live in Alaska. And those that do are very likely to live in Anchorage.

Since Alaska’s population is not growing like the rest of the country, and since our labor force participation rates are much different, controlling the local data with the national data might not be a fair representation of what is really going on here.

Employment Data

The same can’t really be said about the employment numbers. Those are good data coming from credible sources – the employers. The Department of Labor collects this data each month in a report from companies that pay into the unemployment insurance pool.

So, we have pretty good data about how many people have jobs. We also have good data about how many people are collecting unemployment benefits. But it is combined with data that is less likely to be specific to Alaska when they form what we know as the unemployment rate.

For these reasons, I don’t put a lot of stock in the unemployment rate, especially at the local level.

4. Unemployment data is not representative of most of Alaska

And then there’s another problem with data outside of the urban centers of the State. It has to do with what we count as employment and unemployed.

The unemployment statistic is designed to measure the health of a modern western economy. One in which people work at retail stores, manufacturing plants, and fast food chains. One where these employed people take their paychecks to grocery stores, restaurants, and movie theaters.

So we should not be surprised when a population that is not living a western lifestyle has an unhealthy economy by western standards.

Subsistence Lifestyle

No one that understands Alaska would consider people living a subsistence lifestyle to be unemployed. We would point out that they quite literally put food on the table. Yet, they don’t show up in the labor force and no company reports the value they create.

Take that village of 100 people again. We said before that 20 people have “jobs” as defined by the government. But what if 40 others spend their days hunting, fishing, and harvesting? If those people reported themselves as employed, the labor force data would suddenly grow.

Difference in Unemployment Rates Due to Discounted Subsistence

If the national average suggests that there are 10 million people in the country that either became old enough to work, just graduated college, or have decided to reenter the workforce, the model will allocate 3 of those people as living in this village. So, the unemployment rate would get calculated like this:

• 20 jobs reported by the employers + 3 allocated unemployed = 23 person labor force.
• 3 unemployed out of 23 people in the labor force = 13% unemployment rate.

If you counted those other 40 people who are doing work (but not for a wage) the unemployment rate would be just 4.8% (3 out of 63). And who knows if there are actually 3 people looking for “work.”

General Problems

This problem of how we count people’s activities in Alaska is incremental to the problems with the unemployment rate as a statistic everywhere.

For example, if two stay-at-home parents tended to each other’s children and were paid equal wages, they would both be considered employed. But when they take care of their own child, they are not considered part of the work force.

Someone living at home with their parents might spend the whole day playing video games, but might report themselves as unemployed to a surveyor. However, the homeless person on the street doesn’t get asked if they are looking for work at all.

A retired person with a pension and no need to work might answer a survey question that “they would take a job if one was offered.” A person that has been unemployed for over a year might be so discouraged that they stop looking for work, and wouldn’t be considered part of the labor force.

A taxi driver might not report that he is making a living at all, afraid that he may get a tax bill. The daughter of a restaurant owner might not count herself as employed, as she only waits tables for tips.

And a sole proprietor with no clients might report themselves as employed. While an author creating something of value might say the opposite.

The unemployment rate is a tricky statistic.