Auto-enrollment’s impact on plan participation has been monumental. Consider the most recent numbers from large plans administered by Vanguard. According to the firm’s 2017 How America Saves study, adoption of auto-enrollment has increased 300 percent since passage of the PPA, which provided fiduciary relief and tax incentives for implementing the feature.

In 2012, 32 percent of Vanguard plans had adopted automatic enrollment. By 2016, 45 percent had.

That led to an explosion in the number of participants Vanguard serves. In 2012, the firm papered DC accounts for 3.4 million participants. By 2016, it counted 4.4 million participants as clients. Auto enrollment explains the 30 percent increase, as the total number of plans Vanguard administered actually decreased in that time.

In Borrowing to Save? The Impact of Automatic Enrollment on Debt, retirement economists studied the impact of automatic enrollment on Army employees, and found it “significantly” increased car and home loans by 2 percent of income and 7.4 percent of income, respectively.

As non-collateralized debt increased, total retirement contributions were only modestly higher after auto-enrollment —5.8 percent when considering employee and employer contributions.

In 2010, new employees were auto-enrolled in the TSP at a default deferral rate of 3 percent, with a 100 percent match from the Army; the next 2 percent of employee deferrals were matched at 50 percent.

The study shows that the 3 percent automatically deferred to the savings plan was offset by the higher car debt.

The higher home debt further offsets the increased contributions from automatic savings. The economists say automatically enrolled participants are able to obtain larger mortgages by leveraging higher retirement account values.

Auto-enrollment worth more debt?

The good news from the paper is that automatic enrollment is not forcing participants into higher credit card debt. And the paper notes that the higher mortgage debt represents the acquisition of real estate assets that in time will likely increase in value, in turn increasing savers’ net worth.

But the study raises an important question that retirement policy stakeholders will have to weigh: Do automatic savings requirements restrict workers’ cash flow and in turn require other expenses to the financed by higher debt?

While auto-enrollment has succeeded in increasing participation rates, the knock on the design is that it has deflated overall savings rates.

The principle of inertia that supports automatic enrollment—participants will most likely not opt-out of savings plans once enrolled—has been a double-edged sword.

Thanks to inertia, participants are mostly not opting out after enrolled. But neither are they opting to increase savings rates. Vanguard data shows that average deferral rates have decreased as participation rates have gone up.

“The decline in average deferral rates is attributable to increased adoption of automatic enrollment,” write Vanguard economists. “While automatic enrollment increases participation rates, it also leads to lower contribution rates when default deferral rates are set at low levels, such as 3 percent or lower.”

To counter that, plan providers and policy specialists have promoted automatic escalation in accord with auto-enrollment.

Legislation at the federal level proposed by Rep. Richard Neal, D-MA, would require employers to offer 401(k) plans, and set an automatic deferral rate of 6 percent, with an annual 1 percent increase in savings.

And the Oregon Saves auto-IRA program, which will require all employers that don’t offer a 401(k) plan to enroll workers by 2020, sets an automatic deferral rate of 5 percent, with an automatic annual increase of 1 percent.

In both of those plans, participants can opt out of saving, or reduce their deferral rate, much as participants in the private-sector 401(k) plans can.

But as the new research points out, nuanced questions remain as to the practicality of aggressive automatic features.

How much can median earners really be expected to save for retirement? If federal and state policies aim to increase retirement deferrals, how much increased personal debt is acceptable in turn?

And perhaps most pressing: Are lawmakers and industry experts adequately balancing those questions?