Americans are better prepared for retirement than we think.

[clifp]

Today's WSJ has fascinating article on retirement income which high lights one of my pet peeves the awful archaic state of government data collection.

Since it is behind the WSJ pay wall I'll quote extensively from it and provide some commentary..

Quote:

There is a widespread perception that most Americans are inadequately prepared for retirement. The story pushed by pundits and policy makers is that the shift over the past 30 years from defined-benefit pensions to defined-contribution savings plans such as 401(k)s has dramatically reduced retirement income to supplement the benefits provided by Social Security. Sen. Tom Harkin (D., Iowa) has introduced legislation to increase Social Security benefits and build a government-run supplemental saving plan. Sen. Elizabeth Warren (D., Mass.) has so captivated progressives with her demands to raise Social Security payments that she is touted as a potential presidential candidate in 2016. Nevertheless, the story about the declining income prospects of retirees is not true. The story is largely based upon data from the Current Population Survey, conducted annually by the U.S. Census Bureau. Data from the Current Population Survey, or CPS, form the basis of the Social Security Administration's Income of the Aged publication series—which is widely cited as showing that Americans' inadequate retirement incomes force them to increasingly rely on Social Security benefits. But the CPS fails to count most of the income Americans derive from 401(k) and IRA plans, as well as significantly understating the percentage of current American workers who are saving for retirement. The CPS measures the sources and amounts of income received by American households, including income from retirement plans. The Census Bureau's definition of income, however, includes only payments made on a regular, periodic basis. So monthly benefits paid from a defined benefit pension or an annuity are counted as income, while as-needed withdrawals from 401(k)s or IRAs are not... This definition of income obviously understates the resources available to retirees. And this understatement will only grow as more and more Americans save for their retirement through defined-contribution plans and not from defined-benefit plans. The misperception about retirement income becomes clearer when other data are taken into account. For 2008, the CPS reported $5.6 billion in individual IRA income. Retirees themselves reported $111 billion in IRA income to the Internal Revenue Service. The CPS suggests that in 2008 households receiving Social Security benefits collected $222 billion in pensions or annuity income. But federal tax filings for 2008 show that these same households received $457 billion of pension or annuity income.

The rest of the article discuss other discrepancies between the CPS and the IRS data with respect to retirement income.

It is important to note that virtually every report on income, taxes, inequality uses data from either the CPS or the IRS. IMO the IRS is vastly more reliable than the CPS data. This article highlights how inaccurate the CPS data is. (A huge part of the problem is the CPS uses techniques little different from the New Deal when it was first started, long person to person surveys).

The problem with IRS date is it isn't very broad. For instance if you are trying to understand the income of senior citizen, Roth distributions are tax free and the IRS does not collect data on them. Likewise if you are looking at the income of unemployed folks, food stamps are not taxed so the IRS collects no data. God knows how ACA subsidies are treated.

The CPS in contrast collects data on the income from both ROTH distributions and food stamps, but the accuracy is awful.

[LOL!]

And in other news … People continue to retire as they get older.

[jollystomper]

Did the article break things down to an individual level? Just wondering, given the current prevalent philosophy of "some people have too much", if they discussed how this income was spread. One could take the stance that though there are many more retirement dollars that exist but 90% of it is going to 10% of the folks, and that doesn't improve the retirement outlook for the vast majority of people.

[Htown Harry]

The Federal Reserve puts out a useful report...every three years.

FRB: About

Quote:

The regular Survey of Consumer Finances (SCF) cross-sectional surveys are conducted every three years to provide detailed information on the finances of U.S. families. No other study for the country collects comparable information. Data from the SCF are widely used, from analysis at the Federal Reserve and other branches of government to scholarly work at the major economic research centers.

My pet peeve...Investment house reports and news stories that conflate average retirement account value with average retirement savings. Rare is the household that has all of their retirement savings in one account.

[dunkelblau]

Quote:

Originally Posted by Htown Harry The Federal Reserve puts out a useful report...every three years. FRB: About My pet peeve...Investment house reports and news stories that conflate average retirement account value with average retirement savings. Rare is the household that has all of their retirement savings in one account.

Looks like Dr Baker finds that the SCF data is largely consistent with the CPS numbers:
The WSJ Says the Elderly Are Rich But We Didn't Know It | Beat the Press

[rescueme]

I wonder how many of the "1 in 20" post on this board?

1 In 20 US Households Has Over A Million In Assets: This Is Where They Are | Zero Hedge

This being the case, I would say that many (regardless if they hit the "magic number") are ready for retirement.

Just my personal opinion, FWIW ...

[MasterBlaster]

Quote:

Americans are better prepared for retirement than we think

Is anyone really surprised by this ?

Investment houses and brokerages use these biased archaic reports to make people fearful amd to save more so that they can rake in more fees.

Advocates for the poor use these biased archaic reports in an attemt to increase government programs and increase taxes (and debt).

[Animorph]

Just an example of how entrenched "income" is in classic retirement planning. Invest for income, spend the income from a trust. Total return still has a ways to go to overcome this.

[karluk]

I find this WSJ article to lack credibility. It's all well and good to talk about non-periodic withdrawals from retirement accounts as being part of retirees' income, but it's quite another thing entirely to demonstrate that all but the lucky few have any meaningful retirement assets to draw from.

More precisely, what reason do we have to believe that the IRS data, for example, isn't being skewed by extremely high income in a relatively few households? If I take ten retirees, nine of which have no meaningful assets and one of which has $1 million in capital gains income, the IRS data will show that these ten retirees are averaging $100k in retirement income. That makes it sound as if the average retiree is quite well off, when in fact 90% of them would be destitute without their Social Security benefits.

The data I've seen indicate that the top 20% of the households in the country control about 85% of the wealth. That makes the likely retirement income distribution a lot closer to one rich retiree and nine poor ones than a "most retirees are doing well" scenario.

Take a look at the following charts of household weath by quintile. The second through fifth quintiles are definitely not likely to retire with a lot of high balance 401ks to draw from.

Asymptosis » Household Net Worth by Quintile, ’62-’09 (Be Prepared to Scroll)

[clifp]

Quote:

Originally Posted by dunkelblau Looks like Dr Baker finds that the SCF data is largely consistent with the CPS numbers: The WSJ Says the Elderly Are Rich But We Didn't Know It | Beat the Press

That isn't surprising given that SCF uses the same methodology as the CPS, I think it just ask different questions.

Let me illustrate the problem. Most of the panels try and survey the same people over long periods of time. The Census bureau devotes a lot to effort to ensuring the sample is good representation. They even have teams of PHd who devote their careers to develop models to compensate for under or oversampling specific groups. All of which is just dandy.

The problem is when it comes to financial information a ton of American are complete clueless. So for instance one of the panel question is something along the lines of
"over the last 3 months how much dividend income did you receive.
A. less than $100
B. $100-$250
C. $250-$500
d. $500-$1,000
E $1,000-$3,000
F More than $3,000

Now while many forum members could answer that question accurately in response to interview question. I would guess that a large number would need to consult their brokerage statements or tax returns. But if you've already spent an hour taking a survey it is tempting just to guess rather than double check.

However, among the population at large a huge chunk of the population has no idea what a dividend is. They could have $500,000 invested in mutual funds, with capital gains and dividends automatically reinvested. As far as they know they didn't get a dividend so they'd choose "A" <$100, rather than E or F. It doesn't take many folks who answer A instead E to screw up the result badly in trying to determine the income of retirees.

[Gumby]

Quote:

Originally Posted by MasterBlaster Is anyone really surprised by this ? Investment houses and brokerages use these biased archaic reports to make people fearful amd to save more so that they can rake in more fees. Advocates for the poor use these biased archaic reports in an attemt to increase government programs and increase taxes (and debt).

And if I were trying to cut or means test social security, this study would be quite helpful to me.

[Al18]

Quoting Gregory House 'Everybody lies'

I don't trust any of the surveys

I do believe the report stating what the IRS collects and the yearly 401K dollar amounts reported by Fidelity - the largest 401K provider in the US

[nun]

The frighteningly small reported size of DC balances for most Americans makes me question the validity of this WSJ report. Income from 401k etc might be being missed in official statistics.....but how much is that income. I can see how the figures for people who regularly save into a 401k are looking good right now, but what about the people who just don't save. Also no mention is made about people's level of debt.

I take most of these surveys with a pinch of salt. The only ones I take notice of are the one's put out by the Center for Retirement at Boston College.

Center for Retirement Research

[nun]

Here's an interesting study.

http://crr.bc.edu/wp-content/uploads...12-24-5081.pdf

[dunkelblau]

Quote:

Originally Posted by clifp That isn't surprising given that SCF uses the same methodology as the CPS, I think it just ask different questions. Let me illustrate the problem. Most of the panels try and survey the same people over long periods of time. The Census bureau devotes a lot to effort to ensuring the sample is good representation. They even have teams of PHd who devote their careers to develop models to compensate for under or oversampling specific groups. All of which is just dandy. The problem is when it comes to financial information a ton of American are complete clueless. So for instance one of the panel question is something along the lines of "over the last 3 months how much dividend income did you receive. A. less than $100 B. $100-$250 C. $250-$500 d. $500-$1,000 E $1,000-$3,000 F More than $3,000 Now while many forum members could answer that question accurately in response to interview question. I would guess that a large number would need to consult their brokerage statements or tax returns. But if you've already spent an hour taking a survey it is tempting just to guess rather than double check. However, among the population at large a huge chunk of the population has no idea what a dividend is. They could have $500,000 invested in mutual funds, with capital gains and dividends automatically reinvested. As far as they know they didn't get a dividend so they'd choose "A" <$100, rather than E or F. It doesn't take many folks who answer A instead E to screw up the result badly in trying to determine the income of retirees.

When I first looked at those SCF tables I was surprised by the numbers, but they seem tricky to interpret. I think the SCF financial asset holdings in Table 6 on the far right "any financial asset" must be pretty close to the population median because its coverage is over 90% of households for all age brackets. And here we see that the median age 75+ household has: $43500 in total financial assets (as of 2010).

Can this possibly be correct? I assume it's financial assets excluding pensions and social security, but it still seems low to me. The 65-74 age bracket is a bit higher at $71300 median, and the highest age bracket is 55-64 at $77200 median. Lower down Table 6 has a breakdown by percentiles of net worth, looks like overall median is ~40K, and the cutoffs at 75%, 90%, and 95% are at 150K, 500K, and 810K.

So if these SCF numbers are wrong, what would you estimate are the real numbers? Like in your example of $500,000 invested in mutual funds, the SCF data suggests this is only 10% of households (all age brackets pooled). Do you believe this percentage is actually much higher, like 30% or 50%? If I were to guess, I'd say ~50% based on people that I know, but I don't know very many people at this level of detail.

Perhaps we have a really steep upper tail and those who sit on it surrounded by similarly-situated people have a distorted perception of the rest of the distribution?

[nun]

Quote:

Originally Posted by dunkelblau Perhaps we have a really steep upper tail and those who sit on it surrounded by similarly-situated people have a distorted perception of the rest of the distribution?

Someone writing for the WSJ might not have the most complete of perspectives. When the 85 richest people in the world have assets equal to the entire bottom half of the world's population I think there might be a bit of a skew to these financial distributions.

[Walt34]

Quote:

Originally Posted by nun When the 85 richest people in the world have assets equal to the entire bottom half of the world's population I think there might be a bit of a skew to these financial distributions.

Ya think?

I'm sure we all have relatives/friends/acquaintances who will be living solely on SS if they even have that.

[Shawn]

Quote:

Originally Posted by nun Someone writing for the WSJ might not have the most complete of perspectives. When the 85 richest people in the world have assets equal to the entire bottom half of the world's population I think there might be a bit of a skew to these financial distributions.

A related argument is often applied to the NY Times. The NYT staff surround themselves with like-minded people and hence have perspectives quite outside the mainstream. That said, when it comes to income, I suspect that the perspectives of WSJ and NYT writers are those of people with relatively average incomes.

However, I fail to understand how income distribution has anything to do with the merits (or lack of merits) of the article. Yes, there is a difference between mean and median. And yes, it is important to evaluate data by considering all their characteristics (e.g., distribution). But income was skewed 50 years ago too, and people were able to retire. And this was not due to defined benefit plans since these plans were usually small. So there seems to be little correlation between income distribution and retirement.

It is a simple question, although the answer may be complex. Are Americans today better prepared for retirement than is commonly claimed?

Logically, the answer seems to be yes. Despite claims that the average 60 year old has only 18 cents in retirement savings (an exaggeration on my part but not by much), people retire in bulk everyday. Yes, there are people who keep working, but there are also people who retire early. Some by choice, some by circumstance.

It easily could be that people today are financially better prepared for retirement than people in the recent and more distance past. My grandparents on my fathers side retired with minimal savings, SS, and a small military pension. Same with my father and his wife. My grandmother on my mothers side retired with minimal savings, SS, and a very small corporate pension. My mother retired with only about $1K/month of social security. She also has about $85K from the sale of her house but rarely touches these funds. She does not live lavishly, but lives surprisingly well (car, cell phone, computer, etc).

There were no 401k's/IRA's in my grandparents days. Their account balances, and those of their peers, would show up as $0. Yet they were able to successfully retire. The same is true with my parents. So it seems that someone today with only "18 cents" in their 401k or IRA is better off than either my parents or grandparents.

My biggest peeve with these studies is that they usually if not always ignore both taxable savings and multiple retirement accounts. In addition to my taxable savings, which is the bulk of my portfolio, I have 8 retirement accounts. Five of these 8 accounts have balances of less than $100K. The average of these five low-balance accounts is about $50K. Granted, I may not be representative. But these studies may interpret my own data as five individuals with limited savings, instead of one individual who is more than ready to retire.

[mpeirce]

Quote:

Originally Posted by Shawn My biggest peeve with these studies is that they usually if not always ignore both taxable savings and multiple retirement accounts. In addition to my taxable savings, which is the bulk of my portfolio, I have 8 retirement accounts. Five of these 8 accounts have balances of less than $100K. The average of these five low-balance accounts is about $50K. Granted, I may not be representative. But these studies may interpret my own data as five individuals with limited savings, instead of one individual who is more than ready to retire.

Me too.

We have 7 IRA type accounts and even that's after some consolidation. Two of them have substantial balances, while the other's not so much.

Also, our IRA money is not what we are relying on for retirement. I don't plan on touching mine for another 20 years!

Of course, I'm not the "typical person" either, but then who is ;-)

[Hamlet]

One thing to keep in mind is that older people tend to have more wealth than younger people, as time is a key component of wealth accumulation.

When I was just out of college, I had no wealth, but over time I have accumulated decent amounts in my retirement accounts. I'm probably now in that top 20%, or close to it.

So you can't just take the general population numbers and assume that retirees will match that distribution.

People of retirement age are much more likely to be in the top 20% of household wealth than younger people.

Quote:

Originally Posted by karluk The data I've seen indicate that the top 20% of the households in the country control about 85% of the wealth. That makes the likely retirement income distribution a lot closer to one rich retiree and nine poor ones than a "most retirees are doing well" scenario.