How to Budget for Retirement

Before you spend, earn.
Before you invest, investigate.
Before you quit, try.
Before you retire, save.

– William Arthur Ward

The Magic Number

Whole books have been written about how much money is needed to retire. Many websites have retirement calculators that will tell you how much you need to retire. This is the so-called magic number – the net worth you will need to retire. Based on this number, many calculators will even tell you how much you should be saving for retirement or how much you should be contributing to your (subsidized) savings plan. Other apps will calculate your retirement risk. They may calculate the odds that you will run out of money in retirement by a certain age, or they may tell you what investment mix provides the optimal risk for you. In order to properly discuss the magic number, we will need a little background information.

Money Theory

One of the core concepts of finance is the time value of money. The same amount of money is worth more now than it is in the future. If you do not believe that, just ask anyone whether they would rather receive €1,000 now or €1,000 a year from now.

There are several reasons why everyone would prefer it now. First, money that is available now can be spent now. If you have to wait a year, you need to be compensated for the wait. Second, money can be invested to earn interest. Even if you do not plan to spend the money now, you will want to receive the money now so that it can be earning more money for you. Third, ongoing inflation means that this money will not be able to buy as much in a year. Thus, you will want compensation for inflation.

These reasons are interrelated. The interest that you receive by investing is compensation for waiting and for inflation. Hence, we live in a world where interest costs are ever present and embedded into the cost structure of everything.

Money Calculations

Given this time preference for money, we can derive various equations for what money is worth at different times. Discounted cash flow analysis is the process of using the time value of money concepts to value things. Future value tells us how much a certain amount of money will be worth in the future. Present value tells us how much a cash inflow or outflow is worth now. Future cash flows are discounted and past cash flows are increased to reflect the time value of money. Net present value is simply the sum of the present value of all cash flows in question. Businesses use net present value to value projects, loans, companies, and pretty much anything with cash flows.

Spreadsheets and financial calculators contain a collection of related time value of money functions. These functions are simplified versions where the interest rate, cash flow amount, and time between cash flows are constant. While these functions are simplified, they are nonetheless adequate for many common applications, such as mortgages, auto loans, annuities, savings accounts, and bonds.

If you use a spreadsheet like Excel or Google Docs, you will notice the following five standard financial functions:

  • FV = Future Value
  • PV = Present Value
  • RATE = Interest Rate
  • NPER = Number of Periods
  • PMT = Payment Amount

If you specify any four of these five items, the spreadsheet will calculate the remaining item.

Retirement Calculations

In theory, it is easy to calculate the magic number. If spending is constant, we can use a simple present value calculation where we plug in the following numbers:

  • NPER = How long you live
  • RATE = The return on your investments
  • PMT = How much you spend
  • PV = The Magic Number

Even if we want to adjust for inflation, we can just incorporate that into the rate (i.e. an inflation adjusted rate). If the payments are irregular, we can still easily use a spreadsheet to perform discounted cash flow analysis. We will just need to use a few other financial functions. Yes, the math to calculate the retirement magic number is reasonably straightforward.

How Much To Retire?

So how much is needed for retirement? Based on net present value calculations, the answer is zero! The answer is not zero cash or zero assets, but a net present value of zero. This means that if you stop working now and you discount back all future cash inflows and outflows in your life, the net present value must be at least zero. In theory, if you discount all the passive income you have coming to you in the future (i.e. pensions, retirement accounts, dividends, social security, etc.) and you also discount everything you are going to spend in the future (i.e. all living expenses, medical care, etc.) and the net present value is at least zero, then you are all set. In theory, it is simple to calculate how much to retire.

Magic Number Calculations Are Usually Wrong

Unfortunately, theory is not often the same as reality. And if you have not already guessed, the problem is not the formulas, but the data to plug into the formulas. Each of these inputs is almost complete guesswork. No one really knows how long they will live. No one really knows the (inflation adjusted) return on their investments. And no one really knows how much they will spend during retirement.

In our opinion, it is foolish to blindly trust your retirement to a formula for which nearly all of the inputs are crude guesses. The answer is likely to be way off from reality. On the other hand, it is equally foolish to dismiss these calculations as nonsense just because we ca not accurately predict inflation, spending, and longevity. This would be a gigantic mistake. Even though we do not know the appropriate inputs, we can still learn a lot from the model. We can better understand the various retirement risks by understanding the sensitivity of the model to the data inputs.

Retirement Risks

We can start by thinking about what sorts of things can go wrong with our retirement plans. While each person may have a few unique risk items, the most common risks have already been identified and elaborated by others. The Society of Actuaries (SOA) has a very good list of post-retirement risks that can be used as a starting point. We will summarize and bucket these risks into our present value model.

Spending (Payment)

  1. Inflation Risk
  2. Life Event Risk (Death, Divorce, Accident, Illness, Job Loss, Fraud, etc.)
  3. Public Policy Risk (Taxes, Healthcare, etc.)

Investment (Rate)

  1. Investment Risk (Stock Market, Interest Rate, Order of Returns, etc.)
  2. Business Continuity Risk (Company Pension, Company Stock)
  3. Employment Risk (Part-time work during retirement)

Longevity (Term)

  1. Longevity Risk
  2. Early Retirement (planned or forced)

Sensitivity analysis can also tell us the impact of these risks should they arise. An increase in spending is linear. If your yearly spending is going to be 20% higher, then you need 20% more money at the start.

An increase in longevity is actually sublinear. In increase of 20% to your retirement timeline requires less than a 20% increase to your magic number. This risk gets all the press because it is a scary thing to have money at 90 and then run out at 95 “because you lived so long”. But we are not sure that is the most accurate way to look at it. The difference in present value between 30 years and 25 years is small – you only need about 6% more money at the beginning. So we tend to think that if you run out of money at 95, you actually had a really high chance of running out at 90 or 85, but you had thus far been lucky with the ups and downs of the market and/or inflation.

An increase in investment return is perhaps the most critical of the three factors. First of all, the range of possible and likely values is much greater than the other factors. You might live to be 100, but it is pretty certain you will not live to be 150. If you have adequate insurance, your mandatory spending (not including inflation) might increase, but it is unlikely to spiral out of control (except for the obvious exception of healthcare costs). On the other hand, the range of your inflation adjusted investment returns could be huge. It is certainly not out of the question that you could make 20% or even 30% in a given year with essentially no inflation, or you could lose 40% while also taking a 10% inflation hit. And do not assume bonds or cash will protect you because we are talking about inflation adjusted returns.

The difference in money needed for 35 years of retirement versus 25 years is not nearly as great as the difference between inflation adjusted returns of +5% and -5% in the first 10 years of retirement. Frankly, although the chance of a protracted period of inflation is probably greater than your chance of living to age 110, most people do not worry about the former but do worry about the latter. Part of the reason for this is that inflation is not so easy to visualize. However, another reason is that the amount of money one would have to set aside to compensate for a truly bad inflation/market scenario is probably impractical for most people. They would never retire if they had to set aside that much money.

The Elephants in the Room of Retirement Risks

All of the risks discussed above are important. Yet, in our opinion, there are three risks that stand out above all the others.

In third place, we have healthcare risk. Because the costs are potentially very large and the probability is high, healthcare dwarfs many other risks. There are also many variations beyond just having to pay for an extended illness. Medicare premiums may escalate in the future. A serious illness or disability may limit your part-time employment opportunities, or your housing needs may change due to your health. There are a lot of things to go wrong.

In second place, we have a risk that is well known in academic circles, but does not receive much attention in the popular press: the order of investment returns. Note that this is not the same as the variability of returns. Suppose you invest a sum of money for a 20 year period. If you do not make deposits or withdrawals to your investment account, the markets may go up and down, but the order of these fluctuations is irrelevant. If you could somehow take the annual returns for those 20 years, and scramble the order of those years, the end result would still be the same.

However, everything changes if you are making retirement withdrawals from that account. Suddenly the order of returns matters, and in particular, if there is a major downturn at the very beginning, the odds are high that you will not be able to recover from it. Why is that? The problem is that your account could be shrinking significantly while you are also withdrawing from it. One way to look at this is to realize that the money you withdraw for living expenses will never have a chance to rebound later with the market. Another way to think about it is to calculate the withdrawal percentages. For example, if you have a million dollar retirement portfolio and you withdraw 4% of the initial value each year, you will withdraw €40,000 yearly. Now assume that a major market downturn occurs just as you retire. The market gets cut in half over three years – this is always a possibility. At the same time, you keep withdrawing €40,000 each year. After 3 years, your portfolio has shrank to €400,000. Since the market does not know or care about your retirement date, at this point forward, the odds are the same as if you had retired and chosen a 10% withdrawal rate. This is highly unlikely to work. Even if the market strings together a number of +10% years, your reduced portfolio will only tread water, and will likely be wiped out on the next significant downturn. This is a very serious risk.

Finally, in first place, there is inflation. This is the bogeyman that every academic paper and financial advisor warns about, yet it is treated as fiction or a minor nuisance by many people. Please, please, please: take inflation seriously. Einstein declared compound interest to be a wonder, but inflation is like compound interest in reverse. It slowly but ruthless eats away at your purchasing power, and you better take heed if you do not want your retirement derailed by it.

Retirement Strategies

When we read blogs and magazine articles, there seems to be two common retirement planning strategies for dealing with risk: (1) save much more money than you expect you will need, and (2) refine your investments to align with complicated investment models.

We would suggest that the first approach is not really a retirement strategy, but just the idea of being ultraconservative with your estimates. We do not have a problem with that, but for most people, it would not be a realistic strategy. For example, suppose your retirement calculator tells you that you need €2 million to retire, but you do not trust these numbers. So instead of €2 million, you shoot for €5 million or even €10 million. What is so bad about that?

First of all, if the intention is simply to overshoot by a mile, how do you know where to stop? Who will say that €5 million or €10 million is really enough? Second, the average person does not have a healthy appreciation for how hard it is to accumulate just the amount the retirement calculators suggest. In many cases, it will simply not be mathematically possible to shoot for five times the target amount and hope to achieve it on a normal household income. Lastly, there are enormous costs to such an approach. Is it worth delaying your retirement 15 or 20 years, or skimping your whole life to achieve some huge sum of money? We do not mean to be morbid, but clearly there is the very real possibility that you will die or become incapacitated before you achieve it.

As for the second approach, we are again not opposed to the idea. If people want to try to optimize their asset allocation and run Monte Carlo simulations and hedge currency risk and interest rate risk and try to predict future tax policy, we are OK with that. Truth be told, we try to do that ourselves at times. We sometimes wonder, however, if all of that is not just rearranging the deck chairs on the Titanic. We do not have a lot of faith that investment returns and interest rates and inflation and tax policy can be predicted with any accuracy. In fact, we would suggest that if you really think you can predict those variables with any remote degree of accuracy, then you could make a lot more money on Wall Street than you would by optimizing your retirement plan, because most of the predictions from even the highest paid people have been wrong.

Planning For Retirement

Given all that has been discussed thus far, how can we improve upon the typical retirement plan? First of all, it would be helpful to stop blindly plugging numbers into retirement calculators and trying to reduce everything to a single magic number. Secondly, it should be understood that portfolio optimization will only take you so far.

Instead, it is probably more useful to learn how these retirement numbers are calculated and the inaccuracy of these numbers. Then understand the risks involved and how to mitigate those risks. The importance of retirement planning cannot be overstated, but it is better to focus on areas you can control. How to plan for retirement is perhaps less an exercise in mathematics and more an exercise in risk management.

Retirement Risk Mitigation

Here are some of the types of retirement risks we are looking at mitigating. It is not an exhaustive list, and we are not recommending any particular strategy for any particular person. We are listing these ideas to illustrate the kinds of questions we think would be more useful to examine than to simply ask how much to overshoot the results of a retirement calculator.

Order of returns: Do we have a contingency plan to reduce spending or earn income should a major market downturn occur in the first few years of retirement?

Inflation: Is there enough slack in our budget to be able to compensate for inflation by spending reduction or product/service substitution? Could we delay social security benefits since they are indexed for inflation?

Life Events: Do we have adequate insurance for accidents, fraud, and health problems?

Public Policy: Do we have a way of learning about public policy shifts before they become law? Do we have some flexibility in the timing of taxable income? Do we have a mix of taxable and nontaxable income?

Spending: Do we know our current spending level? Have we accounted for employer benefits and depreciation? Do we envision any material changes to our spending in any areas during retirement?

Longevity: Have we considered longevity insurance or annuities?

Investments: Have we reduced any obvious investment risks? Do we have a diversified portfolio? Is the diversification with respect to key risk factors and not just throwing darts at many different investments? Are debt levels low so that debt repayment does not exacerbate investment risks?

Retirement Calculations Across The Blogosphere

In preparation for this article, we decided to read a number of financial blogs, looking for how people were budgeting for retirement. We were a little disturbed at the results. We are sure there are a lot of bloggers who budget and never blog about it, so we were not all that concerned about the general lack of budgeting posts. We were a lot more concerned about the people that did post their budget.

We certainly applaud the fact that some bloggers stress the importance of budgeting to achieve their personal and financial goals. Many of these people are clearly working very hard to keep their budget under control and many of them post their monthly budget. But on the negative side, we did not run across any posted budgets that accounted for depreciation, even though most owned a house and multiple vehicles. We also saw no budgets that accounted for health insurance, presumably because their employer paid for it or perhaps because it was simply deducted from their paycheck before it hit their hands. There was also no budgeting for repairs or for any irregular items of any sort.

But far worse was that fact that some of these bloggers projected their partial budgets into retirement. There were, unfortunately, a lot of claims that mortgage plus utilities plus gasoline plus groceries plus dining equalled a certain low amount of money last month, and so this meant that they would be able to live on this amount in retirement. In our opinion, this is a recipe for financial disaster.

Budgeting For Retirement

While we hope you have found this background information about retirement finances to be informative, you may be wondering why all this information should be in the middle of a series about budgeting. So let me try to explain the connection to budgeting.

  1. The most obvious reason is that saving for retirement begins now.
  2. A less obvious reason is that your budget should be as accurate as possible in order to peg a starting level for your retirement income.
  3. Another less obvious reason is that you may need to spend extra money as you approach retirement to mitigate some of these retirement risks. In some cases, money that is spent on insurance, education, or maintenance may be just as important as your retirement savings.

Final Thoughts

We hope this post has encouraged you to really take charge of all aspects of your retirement. Remember the quote from this top of this article: Before you spend, earn. Before you invest, investigate. Before you retire, save.

Free retirement calculators (bogleheads.org)
Withdrawal methods (bogleheads.org)
Asset allocation (bogleheads.org)
Net present value (wikipedia.org)
List of post-retirement risks (Society of Actuaries, SOA)

Leave a Reply!