But while the calculations are somewhat simple, their results are not necessarily accurate. The outputs from these calculations can impact the decision-making for potential retirees - sometimes for the better, but sometimes for the worse. It is the assumptions within these programs, the unknowns often treated as known, which can cause reality to differ so much from projections. For example, one of the most important determinants for the viability of retirement is yearly expenses during retirement. For decades, the general rule of thumb used by financial planners for income needs during retirement has been that approximately 80% of pre-retirement income will be required to live at the same standard of living after retirement as before. Retirement calculators will often use this figure, or one very close to it. (Some programs give users the flexibility to raise this percentage higher or lower, but doing so would require knowledge that most users probably don't have.)
Projecting your future
If you go to a retirement calculator like this one from TIAA-CREF you are asked your current income and age, existing financial assets, expected future investments, investment style (aggressive, conservative, etc.), your spouse's age and income, the state you live in, and whether you have dependents.
When I entered a hypothetical $60,000/yearly ($5,000/month) income for a 55-year-old, TIAA estimated that this pre-retiree would need to earn about $4,000 a month in retirement. That's was based on the assumption that after retirement he or she would need to live on 100% of their current after-tax income (a 100% after-tax income is roughly similar to an 80% pre-tax income for middle-income wage earners). Based on an estimate of Social Security benefits based on that same income, TIAA estimated $2,119 in Social Security benefits at age 67 (the calculator gives the option of using your own personalized Social Security benefit amount if you know it). From there, you are asked about the amount of your other financial assets, whether they are taxable or not, how conservatively you would invest, and your current rate of additional investment contributions. I listed $200,000 in existing assets in a retirement account invested “moderately” (which by their model, meant 60% of my investments were in stocks). I also said I was saving $500 a month for retirement, or $6,000 a year.
Based on my inputs, and their calculations, it was estimated that I would have an income shortfall of about $500 a month at retirement age, age 67. To close that supposed income gap, the suggestion offered was to save an additional $400 or so a month.
My experience was not atypical. Even with stated assets in excess of the average 55-year-old pre-retiree in America, a contribution rate also higher than average, and with a Social Security check covering half of my projected monthly spending, I was still not saving enough, based on the model's expectations. Having used dozens of these calculator for real individuals contemplating retirement, nearly all I have seen or that has been reported to me, showed that the pre-retiree was not saving enough. In my experience, it is rare that one is reported as saving enough to retire comfortably, even with a high savings rate, significant financial assets and an intention to work to “normal” retirement age, currently age 67 (age one receives full Social Security benefits).
I should note that this calculator by TIAA is one of the better ones I have seen, acknowledging the presence of Social Security income, differences in state tax rates, existing assets, and whether or not the retiree has dependents. Other “retirement” calculators are wildly deficient, in some cases, ignoring outside sources of income or outside financial assets (such as when a 401(k) employee is told he or she isn't saving enough when the program does not even ask about other financial assets the employee holds).
The most likely source of inaccuracy in retirement calculations probably centers on projected retirement income during retirement. (Though there are others, and I'll talk about a few more next time). It is also very possible that these calculators, and the often inflated “necessary” retirement savings calculated for retirees, have scared million of Americans into believe that retiring at youngish age is not possible for them. A survey by GoBankingRates last year found that more than a quarter of American workers believed they would not retire until their 70s or later.
At first glance, and based on what we so often told about the inadequate savings by pre-retirees (sometimes for good reason) it seems that many of today's workers have good reason to worry about retirement or see retirement in their 60s as an impossibility. Estimates say about a third of Americans have little or no retirement savings. The GoBankingRates survey found that a majority (56%) of Americans had less than $10,000 of retirement savings. However, it is also true that those closer to retirement are more likely to have retirement savings, as well as carry higher average retirement balances than those farther from retirement (see two graphs below.
Most American retirees are financing most of their retirement with Social Security
Except for those with very high levels of financial assets, such as several times the median retirement account savings shown above, most Americans will depend on Social Security to a large degree for their retirement. They may not want to live that way in retirement, but that is the reality for most current retirees as well as for the coming generation of retirees. And especially for those that live in lower cost-of-living areas and have a modest standard of living, many live comfortably mostly or solely on Social Security.
According to the Social Security Administration, more than four in ten (43%) of unmarried retirees in America rely on Social Security income for more than 90% of their income, and more than seven in ten (71%) unmarrieds rely on Social Security income for more than 50% of their total income. Half of married couples rely on Social Security income for a majority of their retirement income and about a quarter of them rely on Social Security for more than 90% of their retirement income. Overall, Social Security income makes up a majority of household income for about 60% of retirees, and more than 90% of retirement income for a third of retiree households.
Given the very low interest rates from low risk investments during recent years, it is likely that Social Security incomes will become even more important to retiree's income. Compared to a generation ago, an equivalent amount of financial assets (even after adjusting for inflation) will produce much less income if invested in traditional retirement savings such as Certificates of Deposit. Today, even most of those with relatively large amounts of retirement savings will not produce enough income (invested conservatively) to be completely independent from Social Security income. (Pension income used to also be a very significant component for the average retiree's income, but its declining use over the last few decades has made it much less relevant for most of today's retirees.) The reality is that Social Security income is critically important to the standard of living for most retired Americans. And given current trends, there's no reason to believe it will be any different for the next generation of retirees.
What if you're not average?
The reason that these calculators can be so misleading is that they rely so heavily on averages. Average historical investment returns, average longevity, average inflation, average changes in spending after retirement. But these averages do not account for the variation away from averages and the unique circumstances facing retirees. Some retirees will live just a few years in retirement, others four decades or more. Long-term average investment returns vary widely among individuals, even among those with the same investment style. Some retirees will live on a third of what they earned before retirement, others will spend twice as much as they earned before.
Probably the most important assumption, and source of innaccuracy, across the spectrum of retirement calculators is the amount of income required to live on each year. As noted earlier, the established rule-of-thumb regarding post-retirement expense requirements, is that retirees will live on 80% of before-tax income, or 100% of after-tax income (very roughly equivalent for most of middle America). Interestingly, this figure of 80% of pre-retirement income appears to be inflated, and not corresponding to the actual experience of most retirees. The evidence shows that retirees are, on average, living on closer to 60% of pre-retirement income (see graph below).
Below is a chart from Advisor Perspectives showing median incomes in the U.S. (Note that this is earnings by household, not by individual.) We see that for those over the age of 65, average household incomes are about $39,000 per year. Most of this comes from Social Security income, though about a fifth of those over the age of 65 have earned income as well, which helps to boost that figure.
If we look at the two age groups just prior to age 65, ages 45-54 and ages 55-64, we see that incomes for the latter age group have already fallen compared to the younger, and highest-earning age group, those aged 45-54. This makes sense because we know that the average retiree does not actually retire at age 65 or 67, but earlier, around 63. Early retirement is sometimes voluntary, but often not, most commonly associated with health or personal issues or corporate downsizing. And of course since some are retiring even earlier, say, age 55 or 60, many in this age range will see significantly lower income, pushing down average wages for this age group.
So if we want to accurately compare before-retirement incomes to after-retirement incomes, we should really be focusing on the younger 45-54 age group. As noted, the average (median) income for those households over 65, is about $39,000 a year. And the average income for households in the 45-54 age range was about $77,000. (It's true that those in the upper age groups will have slightly smaller household sizes, but not much less, especially where most of the population is centered - ages 65-85.)
One important thing to note is that retirees over 65 are living on barely half of those in the 45-54 age group, far less than the 80% benchmark that is so often used. This makes sense if we consider some of the changes in spending that often occur when transitioning from pre-retirement to retirement.
For example, most financial calculators do not recognize the impact that debt and home equity has on retirement. The extent of debt held, especially regarding housing, is extremely important to the financial well-being of an individual or couple, and may actually be more significant than other financial assets. A retiree who owns a house free and clear, without a mortgage, will be in a better place financially than one renting or with little or no home equity. The extent of home equity may also affect the cash flow of a retiree, massively changing the necessary income required to live in retirement. For example, a retiree in her late 50s or early 60s earning $5,000 a month and contemplating retirement, may be told that she will need $4,000 a month to live on during retirement (80% of pre-tax income). But if that pre-retiree is currently paying $1,500 a month on a mortgage that will be paid up within three years, such a substantial reduction in necessary spending should be noted and factored into the retirement equation and necessary income during retirement.
A similar impact on cash flow often occurs with college expenses. This is another spending variable that is ignored by most financial calculators. As college-paying parents know, during those college years the necessary costs are generally very unusual and very significant, often leading to extra work or earnings by parents (for example, both parents working full-time to pay the additional college expenses), possibly more debt, less saving, and more spending overall. If for example, a couple is earning $100,000 a year, paying high taxes, a hefty mortgage, and $20,000 a year for college costs for their children, it is illogical to assume the same after-tax income would be required during retirement, when those college costs (and possibly mortgage as well) would no longer be there.
Other debts can impact the future financial picture as well. A couple with high auto loan payments, student loan or other debt payments that forsees a retirement without these or much less of them, could also contemplate a much lower cost-of-living retirement, and one much lower than one suggested by most financial calculators.
In general, the most significant problem with these calculators is how much they ignore the personal needs and circumstance of individual retirees. As a starting point, I think these calculators can be helpful, but the assumptions within them often provide erroneous estimates of future financial needs for millions of pre-retirees. I'll talk about some of the other embedded mistakes within these calculators next time, including the impact that inflated estimates of investment earnings have on future retiree needs.