Thanks to beneficial economic trends between 2013 and 2016,there was a small amount of improvement in the National RetirementRisk Index, which tracks the percentage of working-age households“at risk” of being unable to maintain their preretirement standardof living in retirement.

According to a brief from the Center for Retirement Research atBoston College, the improvement in the NRRI—a drop from 52 percentto 50 percent—came about largely as a result of increasing homeprices, although gains from the stock market also contributed.

The NRRI is constructed using data from the Federal Reserve’s2016 Survey of Consumer Finances, a triennial nationallyrepresentative survey of household finances.

And, says the brief, “Since the last SCF was conducted in 2013,the U.S. economy enjoyed a period of low unemployment, risingwages, strong stock market growth, and rising house prices. Thesefactors should have improved households’ preparedness forretirement.” They did, but only by two percent. And that leaveshalf of American households in a tight spot.

Another less-than-cheerful finding of the NRRI is that thegradual rise in Social Security’s Full Retirement Age and lowinterest rates have acted as headwinds, making it tougher forAmerican workers to manage retirement readiness.

In an effort to see whether the positive economic trends didmore to help Americans in saving for retirement than the negativeones did to hinder them, researchers looked at NRRI results by age,income and pension coverage, then assesses the overallreasonableness of the NRRI’s findings, based on its conservativeassumptions.

Some other studies “conclude that most Americans are savingoptimally to meet their consumption needs in retirement, with lessthan 10 percent of households falling short.” But that’s based ontwo “key” assumptions: first, how children affect replacement ratetargets, and second, how households consume their accumulatedwealth in retirement.

These “optimal” assumptions regard households as havingannuitized their financial assets and the proceeds of a reversemortgage, “when few actually do so.” They also assume thathouseholds reduce their level of spending and increase their levelof retirement savings on their own once their children are grownand gone—although this assumption too is often not the case.

Instead, households often remain at the same level of spendingor only slightly reduce it, and few increase their savings level byas much as the optimal assumptions project. NRRI’s projections arecloser to how American workers actually behave.

After reviewing all the data, the brief concludes that “half oftoday’s households will not have enough retirement income tomaintain their pre-retirement standard of living, even if they workto age 65 and annuitize all their financial assets, including thereceipts from a reverse mortgage on their homes. Instead, the studyindicates that “many of today’s workers need to save more and/orwork longer to achieve a secure retirement.”