Are Ninth District states’ unemployment trust funds ready for a downturn?

At the beginning of 2001, the state of Wisconsin had a healthy $1.8 billion trust fund to help unemployed workers make ends meet during their job search. Then, through two recessions—one that year and the Great Recession that followed seven years later—the fund was drained, and the state had to borrow from the federal government.

“We started with our highest trust fund balance ever,” said Thomas McHugh, who oversees the accounting bureau in Wisconsin’s Unemployment Insurance Division. “By the end of the decade, we were in debt.”

Employers in the state paid $370 million in additional taxes to repay the loan with interest, and state taxpayers chipped in another $25 million in interest, he said.

On the other end of the Ninth District, Montana also started 2001 with a sky-high trust fund of $181.7 million, but saw it shrink by more than half in the Great Recession. Montana, though, bounced back as strong as ever because policymakers had passed a law decades ago to automatically adjust taxes and benefits to economic conditions.

During this volatile period, of the six states in the district, Wisconsin and three others depleted their unemployment trust funds and were forced to borrow to meet obligations to jobless workers. Indeed, around the nation, two-thirds of states did the same, forcing the federal fund that lends them money to itself borrow from another fund.

Today, Ninth District states enjoy comfortably low unemployment rates during what may soon be the longest economic expansion in U.S. history. States have had time to prepare their trust funds in the event of an economic downturn.

According to the latest U.S. Department of Labor report, four of the six states are set with trust funds at or higher than solvency level, meaning at low risk of being drained by high unemployment. (See “Snapshot: A state-by-state look at solvency.”)

On the other hand, Wisconsin and Minnesota are among those with funds still at risk. Michigan’s fund is technically above solvency, but the state got there after swapping federal debt, which counts against solvency, with private debt, which doesn’t.

Measuring solvency

Unemployment benefits are treated as a kind of insurance under U.S. law, one in which employees are beneficiaries and employers pay the premiums based on their history of layoffs. Every state, plus the District of Columbia, Puerto Rico, and the U.S. Virgin Islands, offers its own unemployment insurance paid out of its own trust fund. The federal government oversees all of the individual state funds (See “A primer on unemployment insurance.”)

Many economists see this guaranteed benefit not just as a safety net for workers but as a way to take the edge off recessions. Jobless workers tend to spend most of what they get, stimulating the economy and forestalling further job loss. When states cut benefits in response to insolvency, as Michigan did when it cut the maximum weeks of benefits by six, they can make a bad situation worse.

But, as with all insurance, there is a risk that economic catastrophe will lead to such high payouts that a state fund will run out of money. And the magnitude of the challenge is huge: During the peak years of unemployment in and after the Great Recession, unemployment insurance nationwide paid out $198.5 billion, adjusted for inflation.

To measure that risk of insolvency, the federal government uses an average high-cost multiple, or AHCM. The government takes the average of the three highest payouts of unemployment benefits each state paid over the past 20 years and determines whether the state has enough in its trust fund to pay that level of benefits for a year under current conditions.

This rating means the amount of money needed for solvency relative to the size of the workforce can be dramatically different because each state’s economy reacts differently in recessions.

For example, Michigan, which tends to have very high unemployment—surpassing 10 percent in the 1980s and the Great Recession—needed about $1,200 per insured worker in its fund to be solvent as of 2017. South Dakota, which tends to have low unemployment—annual unemployment exceeded 5 percent only a couple of times, in 1982 and 1983—needed just $230.

On the eve of the last recession, Wisconsin was among the least prepared states with its trust fund at only 29 percent of solvency. That meant the state had enough money on hand for three and a half months of benefits if unemployment were comparable to the three worst years. In the Ninth District, only Michigan had a lower solvency; its trust fund was already depleted from several years of economic distress, and the state owed money to the federal government.

Wisconsin’s challenge

Wisconsin’s unemployment insurance system, which state lawmakers approved in 1932, served as the model for today’s nationwide system. (See “Unemployment insurance’s roots lie in Ninth District.”) For decades, the state’s trust fund ran in model fashion, shrinking during recessions and growing during economic expansions.

The exception was the severe recessions of the early 1980s, when unemployment exceeded 10 percent. Despite starting the downturn above the solvency level, the fund emptied, and the state was forced to borrow from the federal government.

Wisconsin repaid those loans within four years, before interest kicked in, and then rebuilt the funds, exceeding solvency by 29 percent at one point. The funds even managed to earn enough interest through the 1990s to keep taxes lower than the amount of benefits paid out, according to McHugh.

After the 2001 recession, the fund again shrank, but the state didn’t rebuild it and was forced to borrow during the Great Recession. State leaders then chose not to repay the federal loans as quickly as possible, causing federal unemployment taxes to effectively go up for all employers in the state.

Minnesota’s reform

At the heart of the Ninth District, Minnesota was also less prepared for the last recession with its trust fund at 38 percent of the solvency level at the end of 2007, enough to pay for just four and a half months of benefits.

The state has had to borrow from the federal government frequently in recent history, with loans on the books every year but one from 1975 to 1985. Unlike Wisconsin, Minnesota failed to rebuild its trust fund to solvency during the economic boom of the 1990s. When the 2001 and 2007-09 recessions hit, Minnesota again was forced to borrow.

Rick Caligiuri, the head of Minnesota’s Unemployment Insurance Division, said the state reformed its unemployment taxes before the Great Recession and, in the recovery, those reforms allowed taxes to go up quickly to replenish the trust fund before dropping just as quickly to give employers relief.

By 2015, the loans to the feds were repaid, and the trust fund enjoyed a whopping 106 percent of solvency, achieving solvency for the first time in 45 years.

The flip side was that businesses started complaining that the high balance showed taxes were too high, causing lawmakers to cap the fund at just above solvency.

Montana’s model

In the aftermath of the last recession, Montana saw unemployment spike to its highest level in two decades. The amount of benefits the state paid out doubled through the Great Recession as unemployment surged. This acceleration was actually greater than what Wisconsin and Minnesota experienced and, in theory, it should have been a shock to Montana’s system.

But it wasn’t. Montana’s trust fund never fell lower than 70 percent of solvency, the lowest in two decades, and it bounced back to above 100 percent within two years. As of the last solvency report, Montana was at a solid 151 percent of solvency. This is the norm for the state, which has kept its trust fund above 100 percent every year for 26 of the past 30 years.

Montana learned its lesson in the recessions of the 1970s and 1980s, when it depleted its trust fund and was forced to borrow. In 1985, state lawmakers passed legislation to ensure that there would be no repeat. They raised taxes on businesses by automatically adjusting not just the tax rate but also the taxable wage, the portion of a worker’s pay used to calculate taxes, so that it goes up as average wages in the state go up. This is known as indexing the taxable wage base.

Montana lawmakers also added a two-year 0.3 percent surtax to pay off the federal debt.

Despite the higher taxes, the law received broad support. An advisory council that included many business groups had proposed it.

One simple trick

The indexing that Montana did was something unemployment insurance experts have advised all states to do for decades to reach solvency. Unfortunately, most states that saw trust funds depleted in the 1970s and 1980s didn’t respond the way Montana’s policymakers did.

A taxable wage base is a concept similar to the taxable income that’s subject to federal income tax. North Dakota’s base, for example, is the first $36,400 of a worker’s wage. For a worker earning $36,300, the tax rate is applied to all of it, but for a worker earning $36,500, the tax rate is applied to just that base of $36,400.

By design, unemployment insurance is meant to tax an employer more if it’s responsible for layoffs that lead to more money being drawn from the trust fund. When the base is very low relative to wages, taxes on all employees earning more than the base are the same, even if one employee earned a much higher wage than another and would receive higher benefits. Low bases effectively subsidize employers that have a potentially higher impact on the trust fund.

The federal government itself has kept the same taxable wage base of $7,000 since 1983, which provides the floor for all states. Nearly every state is higher, but many adjust their bases only occasionally rather than yearly as is done with indexed bases such as Montana’s.

Wisconsin has never indexed its base, but its lawmakers used to make frequent updates until 1986, when the base froze at $10,500 for two decades. The base has seen more updates since the Great Recession, although the current base of $14,000 is still less than what it was in 1986 in real terms. Adjusted for inflation, the 1986 base would be worth $24,000 today. South Dakota and Michigan also don’t index, and the latter’s $9,000 base is the lowest in the Ninth District.

Minnesota, Montana, and North Dakota all index bases to their average wages. That’s made a difference. When Montana started indexing in 1986, for example, its base was $12,200. That’s $27,900 in today’s money. The current base is $33,000.

Targeting solvency

Another simple policy fix experts advise is for states to make solvency their goal and set tax rates accordingly. The federal government has enshrined this goal in a new rule. Starting this year, a state trust fund can only receive federal loans with no interest if it reaches solvency at least once in the past five calendar years.

Many states continue to target a fixed dollar amount that is disconnected from current economic conditions. In Wisconsin, tax rates increased only until the fund reached $1.2 billion, which is well below today’s solvency level. State lawmakers set this target in 1998 when a dollar was worth 35 cents less.

Other states set goals that are simply not in line with historic conditions. In Michigan, tax rates go down automatically when the trust fund is 1.2 percent of the total wages of insured workers. That doesn’t jibe with the state’s three worst years, the statistic used to measure solvency, when total benefit payout averaged 2.17 percent of its trust fund.

North Dakota and South Dakota both target solvency with their tax rates, while Montana targets a percentage of total wages, as Michigan does, but that percentage is much higher than the payout rate in its three worst years.

These policy fixes don’t make trust funds recession-proof, but they do seem to help. Through the Great Recession, of 23 states with funds below solvency level at the start of the recession, more than 80 percent ran dry and were forced to borrow, with Michigan, Minnesota, and South Dakota among them. Of 30 states with funds at or above solvency when the recession began, less than half were forced to borrow.

Of 36 states that didn’t index their taxable wage bases, more than 80 percent were forced to borrow, Michigan, South Dakota, and Wisconsin among them. Of the 17 that did index their bases, only 40 percent were forced to borrow, Minnesota among them.

30 years of warnings

The difficulties states have had with reforms are echoes of the past.

When Congress created today’s unemployment insurance program during the Great Depression, it acted because many states had tried on their own and failed.

The federal government has urged states to bring trust funds to solvency levels for decades. In 1988, after three devastating recessions, the then General Accounting Office warned Congress that the “Unemployment Insurance system has inadequate reserves, and many state trust funds will likely be unable to pay benefits in a future recession without multibillion-dollar borrowing.”

While the recessions had been severe, the GAO said state policies that increased benefits by tying them to inflation but didn’t increase taxes made things worse.

Fast-forward to 2010 and the GAO, now the Government Accountability Office, had similar complaints in another post-recession report to Congress: “While benefits over this period have remained largely flat relative to wages, employer tax rates have declined.”

The agency expressed surprise that during the 10-year economic boom leading up to the 2001 recession, many states failed to build their funds to the solvency level.

In June, the current economic boom will have exceeded that earlier expansion. Still, 29 state trust funds remain below solvency, and many haven’t enacted reforms that could help.