Retirement Income

Decoding Retiree Spending

March 22 2021

A better understanding of spending patterns may transform retirement income solutions

Key Insights

  • Income replacement strategies often assume inflation-adjusted spending in retirement is flat; however, the data reveals that retiree spending declines annually by 2%, and it varies by wealth.
  • Retirees typically choose to adjust their nondiscretionary spending (often considered fixed spending) to match their guaranteed income, challenging the notion that these expenses are truly fixed.
  • Retiree spending behavior reveals a preference for asset preservation. Aligning products and services that account for this will hasten adoption of retirement income solutions.

Understanding how retirees make their spending decisions is crucial to creating retirement income strategies. Studies show that retirees spend less as they age. But why they spend less and how they determine how much to spend is less well understood. Fear of running out of money could be a motivating factor, but what else influences how much people spend?

Employers and plan sponsors need to understand the motivations behind retiree spending in order to provide optimal retirement income solutions. Defined contribution (DC) plans are quickly becoming the major source of wealth for retirees. For some, this may be the largest amount of money they have, and this is where they will be taking distributions to cover retirement expenses. Employers, as sponsors of DC plans, are a key party to this process. Accordingly, many employers are beginning to think or rethink about what role they can or should play in providing retirees with tools to help draw down their DC plan assets in retirement.

To better grasp what drives retiree spending, T. Rowe Price used data from a longitudinal panel of 1,470 households interviewed between 2001 and 2015 from the Health and Retirement Study (HRS)1 and its supplement, the Consumption and Activities Mail Survey (CAMS).

Employers and plan sponsors need to understand the motivations behind retiree spending in order to provide optimal retirement income solutions.

— Sudipto Banerjee, Ph.D. Vice President, Retirement Thought Leadership
Steady Decline in Retiree Nondiscretionary Spending Drives Down Total Spending

(Fig. 1) Inflation-adjusted total household spending declines by 2% annually after age 65

Inflation-adjusted total household spending declines by 2% annually after age 65

Source: T. Rowe Price estimates from Health and Retirement Study (2001–2015).

Nondiscretionary Spending:

Mortgage, rent, utilities, homeowners’ or renters’ insurance, property taxes, home repairs and maintenance, housekeeping supplies, auto payments, auto insurance, auto maintenance, clothing and apparel, health insurance (including supplemental insurance), prescription and nonprescription medication, health care services, medical supplies, food and beverages (excluding dining out), gasoline.

Discretionary Spending:

Trips and vacations; household furnishings and small equipment; charitable and political contributions; cash or gifts to family or friends; dry cleaning and laundry services; home cleaning services; supplies and services for gardening and yard; personal care products and services; tickets to movies, sporting events, and art performances; gym and other sports activities; hobbies and leisure equipment; dining out and takeout food.

Source: See Appendix A, p. 7 for more details. Retirement Study (2001–2015).

Spending Changes With Age

Conventional thinking believes that nondiscretionary spending—basics like housing, food, and utilities—is fixed and does not change over time. However, if we look closer, most of us do have some choice over how much we spend on these items. For example, one might move from New Jersey to Arizona to lower the housing costs or buy groceries in less expensive stores. While housing and food remain indispensable spending items, there are ways to live or shop more economically.

The data shows that the decline in overall retiree spending is primarily driven by a decrease in nondiscretionary spending. Looking at how median annual household spending changes between ages 65 and 90 (Fig. 1), we found that inflation-adjusted or real spending declines at an annualized rate of 2%.

Reduced Spending in High-Net-Worth Households Is Driven by Reduced Nondiscretionary Spending

(Fig. 2) Median household spending declines 2.7% annually for the top 20% of net worth

Median household spending declines 2.7% annually for the top 20% of net worth

Source: T. Rowe Price estimates from Health and Retirement Study (2001–2015).

Low-Net-Worth Households’ Spending Is Flat and Consists Mostly of Nondiscretionary Expenses

(Fig. 3) Median household spending declines by only 0.3% annually for the bottom 20% of net worth

Median household spending declines by only 0.3% annually for the bottom 20% of net worth

Source: T. Rowe Price estimates from Health and Retirement Study (2001–2015).

Spending Is Personal and Varies by Net Worth

While spending goes down for most retirees as they age, the extent of this decline varies widely. Spending is primarily a function of one’s wherewithal, so we looked at how spending varies by net worth (defined as financial assets including value of a primary residence less all debt including a primary mortgage).

We examined the annual spending trends for households in the top 20% (Fig. 2) and bottom 20% (Fig. 3) of net worth. These households had a net worth of more than $667,000 or less than $37,000, respectively. Two different patterns emerge.

  • Spending for the top 20% drops rapidly after 65, remains mostly flat between 70 and 80, and then declines rapidly again. However, spending for the bottom 20% remains largely flat.
  • Median spending drops at a much faster rate for those in the top 20%. The annualized decline in real household spending is 2.7% compared with only 0.3% for the bottom 20%.

Retirees Adapt Spending to Their Income

(Fig. 4) After age 65, the ratio of nondiscretionary spending to guaranteed income quickly approaches 1:1

After age 65, the ratio of nondiscretionary spending to guaranteed income quickly approaches 1:1

Source: T. Rowe Price estimates from Health and Retirement Study (2001–2015).

High-Net-Worth Retirees Can More Easily Align Nondiscretionary Spending to Guaranteed Income

(Fig. 5) The ratio of nondiscretionary spending to guaranteed income veers further from 1:1 for low-net-worth households

The ratio of nondiscretionary spending to guaranteed income veers further from 1:1 for low-net-worth households

Source: T. Rowe Price estimates from Health and Retirement Study (2001–2015).

Fixed (Guaranteed) Income:

Social Security income, defined benefit pension income, and income from any other annuity.

Source: Based on income data from RAND HRS data files.

Asset Preservation Drives Retirement Spending

Conventional retirement income planning assumes that retirees want to maintain a certain standard of living or a certain level of spending and attempt to generate enough income to support that spending level. But the data suggests that the opposite might be true. People are flexible about their spending and adjust it to match their income so that they can avoid drawing down their assets.

Figure 4 shows that between the ages of 65 and 70, the ratio of nondiscretionary spending and guaranteed income drops sharply toward 1 and after age 70 it remains remarkably close to 1, where nondiscretionary spending equals guaranteed income.

What does this tell us?

After a short adjustment period, we see that retirees set their nondiscretionary spending to match their guaranteed income, rather than choose their guaranteed income to match their nondiscretionary spending. This behavior suggests that retirees are reluctant to bridge any spending gap (between nondiscretionary spending and guaranteed income) by buying additional guaranteed income such as annuities.

The most likely reason behind such behavior is wanting to preserve existing assets. Most retirement spending models and retirement income strategies assume that retirees are OK with spending down their assets.

But if retirees do not want to dip into their nest eggs on an ongoing basis for their regular spending, the first step of any retirement income strategy will be to understand why retirees choose to preserve their assets. Whether it is a behavioral inability to spend down, a fear of high expenses late in life, a mix of both, or something else. This understanding will let us provide the right retirement income products.

The behavior of higher-net-worth retirees disputes the argument for additional annuity income.

— Sudipto Banerjee, Ph.D. Vice President, Retirement Thought Leadership

After age 65, retirees in the top 20% of net worth (Fig. 5) quickly adjust their nondiscretionary spending and match it to their guaranteed income, keeping the ratio close to 1. Retirees in the bottom 20% of net worth (Fig. 5) struggle to do that and the ratio veers further from 1 and remains mostly above 1 throughout retirement. This means their nondiscretionary spending exceeds their guaranteed income and suggests these retirees are already stretched thin and are unable to absorb any financial shock. So, even if these less wealthy retirees are highly annuitized through Social Security, they might benefit from additional annuity income.

The behavior of higher-net-worth retirees disputes the argument for additional annuity income. Despite their relatively high wealth, they choose to reduce their spending to match their guaranteed income. This shows that they have enough room in their budget to adjust if their income drops and they might not want to lock up their money in annuities. Accordingly, products such as managed payout solutions might be more desirable for these wealthy retirees because they would enhance the predictability of income without sacrificing control and ownership of their assets.

If asset preservation is the key force driving retirement spending, what is driving that need? If it’s the fear of the unknown—like an unexpected large nursing home bill—then appropriate financial products (i.e., long-term care insurance) can address that. However, if asset preservation is driven by behavioral factors such as an inability to switch from a saving to a spending mentality or an aversion to declining account balances, the retirement income problem becomes harder to solve. The reality is, there is a good chance both types of factors are likely to be in play. That’s why it’s so hard to solve the retirement income problem.

Our research suggests:

  • Retirees should express their preference between asset preservation and spending preservation and the reasons behind such preference while building a retirement income strategy with their financial professional—if they have one.
  • Financial professionals should consider their clients’ personal preferences and financial flexibility when formulating their financial plans. They should also regularly revisit their clients’ retirement income plans, particularly for higher-net-worth clients.
  • Employers and plan sponsors should offer a variety of solutions (e.g., managed payout funds, managed accounts, dynamic qualified default investment alternatives, annuities) and, if possible, professional guidance to help their workers make informed choices about their retirement income needs. Emphasis should be put on products and services that are consistent with observed retiree behavior.

Providing Relevant Retirement Income Solutions

The data presented indicates that, contrary to conventional wisdom, retirees have some flexibility in choosing how much to spend on their nondiscretionary items and that they limit their spending on them to their guaranteed income. This implies that if there is a gap between guaranteed income and nondiscretionary spending, retirees’ first response might be to cut back on nondiscretionary spending, rather than find a solution that allows them to keep spending at the same level.

The challenge is that cutting back spending won’t work for all retirees. There comes a point where it’s impossible to spend less. This is often referred to as a floor consumption level, and it affects people of fewer means. That’s why less-wealthy retirees will struggle to keep their nondiscretionary spending below their guaranteed income and are more likely to benefit from additional fixed or guaranteed income. Yet, it is premature to assume that they will buy annuities if offered, because preserving the small levels of liquid assets they have will also be a top priority for them. On the other hand, managed payout solutions rather than annuities might be more desirable for wealthy retirees.

This also suggests that there isn’t a one-size-fits-all solution. The needs of wealthy retirees are different from the less-wealthy, and different income solutions will appeal to different groups. Therefore, a solution that combines service with a range of products may have wider utility than a sole product-based solution.

So, how do financial professionals and plan sponsors meet the income needs of current and future retirees?

Employers are facing the challenge of how to meet the varying needs of all their retiring employees. There is a wide range of retirement goals that are often further complicated by the needs of their spouses or partners, various levels of risk tolerance, and different personal circumstances.

There are potential benefits to these solutions, including:

  • Adopting retirement income solutions would transform a retirement savings plan into a true retirement plan.
  • Offering in-plan solutions for retirees benefits all by potentially reducing investment management costs.

In turn, plan sponsors and the financial professionals that serve them will benefit from having workers better prepared for retirement with solutions that more closely align to their needs.

Like any plan design or investment decision, the implementation of a retirement income solution in an ERISA-covered DC plan may trigger fiduciary responsibilities that require careful analysis and documentation and, often, the use of experts. If plan sponsors want to adopt new products and features, they should exercise due diligence in their goal to maximize outcomes for retirees.

Appendix A

Nondiscretionary and Discretionary Spending

The Consumption and Activities Mail Survey (CAMS) collects information on six durable and 33 nondurable spending categories. Durable spending (such as a refrigerator or washing machine) is infrequent, and the use of durable goods is spread over many years. So, it is not useful to look at durable spending in relation to annual income. Therefore, we consider only nondurable spending and divide it into discretionary and nondiscretionary spending.


We use data from the Health and Retirement Study (HRS)and its supplement Consumption and Activities Mail Survey (CAMS). CAMS started in 2001, and we used data from 2001 through 2015. Income data corresponding to each CAMS wave are used from HRS.

More importantly, we used data only from the original CAMS cohort first interviewed in 2001 and then every other year since then. Therefore, this is a panel data analysis as we follow the same group of retirees from 2001 to 2015.

However, we used some filters to make the data suitable for analysis. First, to ensure the integrity of the panel, we kept households that were observed in at least two consecutive surveys. Second, to contain spending outliers we dropped households whose total household spending exceeded three times their total household income in any given year, and change in household spending between two consecutive waves did not exceed 50% in absolute terms. Our final analysis sample consisted of 1,470 households.

All spending and income numbers were inflation adjusted using the consumer price index and presented in 2019 dollars.

So what does all this mean?

Download the full article to learn more and uncover opportunities for Plan Sponsors and Financial Professionals.

Key Findings Article

How the Coronavirus Is Affecting Retirement Saving Full Article

Key Findings Infographic

How the Coronavirus Is Affecting Retirement Saving Infographic

Health and Retirement Study, public use dataset. Produced and distributed by the University of Michigan with funding from the National Institute on Aging (grant number NIA U01AG009740). Ann Arbor, MI.

Important Information

This material is provided for general and educational purposes only and is not intended to provide legal, tax, or investment advice. This material does not provide recommendations concerning investments, investment strategies, or account types; it is not individualized to the needs of any specific investor and not intended to suggest any particular investment action is appropriate for you, nor is it intended to serve as the primary basis for investment decision-making.

The views contained herein are those of the authors as of March 2021 and are subject to change without notice; these views may differ from those of other T. Rowe Price associates.

This information is not intended to reflect a current or past recommendation, investment advice of any kind, or a solicitation of an offer to buy or sell any securities or investment services. The opinions and commentary provided do not take into account the investment objectives or financial situation of any particular investor or class of investor. Investors will need to consider their own circumstances before making an investment decision.

Information contained herein is based upon sources we consider to be reliable; we do not, however, guarantee its accuracy.

All investments involve risk. All charts and tables are shown for illustrative purposes only.

T. Rowe Price Investment Services, Inc., distributor, and T. Rowe Price Associates, Inc., investment advisor.

© 2021 T. Rowe Price. All rights reserved. T. Rowe Price, INVEST WITH CONFIDENCE, and the bighorn sheep design are, collectively and/or apart, trademarks of T. Rowe Price Group, Inc.

Tap to dismiss

Preferred Website

Do you want to go directly to the Financial Advisors/Intermediaries site when you visit ?