Working for the federal government conditions people to think about retirement as happening at a certain point, usually at or after one of the combinations of age and years of service that allow for collecting an immediate annuity.

Retirement planning often involves pulling in a second factor, achieving an income target. A common goal is 80 percent, or even less, of pre-retirement income to reflect decreases in certain expenses, primarily those related to working—such as commuting, purchase of more formal clothing, lunches and so on.

But the best-laid plans often go awry, as underscored by two recent reports from the TransAmerica Center for Retirement Studies and the Employee Benefit Research Institute.

One examined expectations of a date. Retirement planning should account for the possibility that the day will come sooner than expected, because that is far more common than it coming later than expected.

In a report based on a survey and individual questioning, it found that 60 percent of retired sooner than they had planned while only 7 percent retired later. Among the former group, two-thirds did so for employment related reasons such as organizational change, job loss, unhappiness with the job or the offer of a buyout. Of the rest, most did so for health reasons and the remainder for family responsibilities such as becoming a caregiver.

Of those who delayed retirement, three-fifths did so for financial reasons or the need to benefits, although almost all of the rest said they delayed because they enjoyed working.

Today’s older workers are far more likely than those of the past to view retirement not as an abrupt end of working but as a transition from full-time to part-time work or to a different kind of work, it added.

However, it said, there’s also often a disconnect between plans and reality when it comes to work after retirement: above half of those age 50 and older who are working plan to continue working at least on a part-time basis after retiring, but among actual retirees, only 5 percent are currently working at all.

The EBRI report meanwhile showed that your decrease in spending in retirement might not be as steep as you’re counting on. It tracked household spending patterns in the first six years after retirement and found that the median decrease—the point where half were above and half below—was only 5.5 percent in the first year. That hit 12.5 percent by the third or fourth year but the spending reduction slowed down after that.

Further, in the first two years of retirement, nearly half of households spent more than what they had spent just before retirement—28 percent spent at least a fifth more. This declined to a third by the sixth year of retirement. Such households were spread across all income levels, it added.

Looked at another way, in the first two years of retirement, four-tenths of households spent less than 80 percent of their pre-retirement spending, although by the sixth year slightly above half did so.

It said the steepest drop in spending was for transportation, reflecting in part the savings on commuting; that dropped by a quarter in the first two years but flattened off after that.

Does all this mean you shouldn’t even bother to make a plan for your retirement? Of course not. But maybe you’ll want to have more than one.