Sunday, November 16, 2014

The Household Balance Sheet

The household balance sheet is one of many topics I've never quite gotten around to, but New Jersey retirement planner Michael Lonier recently offered to write a guest post and he has provided an outstanding introduction to the topic. Please check out Mike's impressive bio, too.

Enjoy!

Finding the Upside on the Household Balance Sheet

Inescapably the time comes to consider that spending from savings is different from saving from income. You survey your savings and investments, scattered across a number of accounts, and the question arises, how best to manage this money to get the best lifetime outcome for the household when you are no longer earning a high income from full-time employment?

Maybe you’re thinking about rolling over 20 or 30 years of 401(k) contributions made to an assortment of funds almost randomly picked over the years, or are holding mostly cash after exiting the market during the 2008 crisis and now realize your money needs to work harder.

How do you organize—allocate—your savings and investments? How much should you invest in stocks and bonds and keep in cash? How do you decide what is best for you?

The Answer to the Puzzle is More Puzzles

You can go with rules of the thumb, like the old chestnut “age in bonds,” possibly with +/- some number adding a veneer of math to what is essentially an arbitrary amount. You can put your money in a target date fund and pay a fund company .20% to 1.00% to use their proprietary “glide path,” which adds a costly layer of research to the age in bonds formula. (Hint: Index TDFs are cheaper.) Dig deep enough, and your head might spin over how highly researched glide paths can be so different.

You can ask an investment advisor who will give you a short quiz and categorize you as conservative, moderately conservative, moderate, moderately aggressive, or aggressive, and plug you into a model portfolio that has somehow been matched with these categories which have somehow been matched to the quiz answers you gave. If that doesn’t seem right, there are advisors who will give you a longer quiz, or more categories, like south-by-south-west, or both. And whichever direction you turn, you’ll pay maybe 1.00% a year.

Everyone has an answer to your puzzle. It’s just that they’re different answers.

If you want to figure it out yourself, you are encouraged by most experts to think deeply about your risk tolerance, how you are likely to react when the bottom falls out, how much you can afford to lose (none!), and then pick a number between 0% and 100% for stocks based on…your best guess.

Look Beyond Your Investments to Your Balance Sheet

Your savings and investments are just one part of the puzzle, one part of the resources you need to manage to get the best outcome. You need to start at a place that accounts for all of your resources, that balances what you have and what you need, revealing the weaknesses—the risk exposures—you need to overcome. Financially, the place to start is with your household balance sheet.

You are probably familiar with a simple net worth statement that sums up current financial assets and debts and provides a snapshot of solvency. A full household balance sheet for planning goes further than that. The asset side of the balance sheet includes the current value of savings and investments (financial capital), the discounted present value of expected future earnings (human capital), and the discounted present value of expected Social Security benefits and pensions (social capital). The liability side includes the discounted present value of all expected future expenses, including debt service and payoffs.

I won’t cover the details of how to calculate these present values in this post, but though it sounds intensely complicated, it’s not. It’s within the reach of anyone familiar with using Excel, and is no more involved than examining your conscious for risk and divining an asset allocation from an arbitrary formula. More importantly, it represents your specific household situation in a logical and mathematical fashion without making any leaps of faith about theoretical risk, personal inclinations, or market behavior. More math, less magic.

Let’s look at a household balance sheet constructed in this way (below). What does it tell us?


The assets on the left include $1,824,700 of financial capital (FC), which is the current value of the household savings and investments in retirement and taxable accounts. The relatively low $274,800 of human capital (HC) suggests in this instance a household nearing retirement with just a couple years of earnings remaining. Someone with ten or fifteen years of employment ahead of them might have human capital over a million dollars. Not surprisingly, the $1,977,000 present value of social capital (SC)—Social Security and pension benefits over a 30-year retirement plan—is the most valuable asset on this particular household balance sheet.

On the liability side, the present value of all expected future expenses over the life if the plan (about 33 years in this case) is shown as $3,381,400. Subtracting liabilities from assets, the balance sheet shows a cushion or surplus of $695,200.

The financial lifecycle is the process of converting human capital into a stream of income that covers ongoing current expenses while building FC and SC to cover future expenses when HC has been “spent down” going into retirement. The balance sheet shows the current relationships between savings and investments, expected future earnings, accrued social capital benefits, and expected future expenses, all either in current or discounted to current dollars. It is the definitive household financial lifecycle scorecard, and so is enormously useful in answering the question about how best to allocate household resources.

Some simple math allows us to focus on the puzzle of puzzles, the allocation of financial capital (below).


If we subtract the sum of HC and SC ($2,251,900) from liabilities ($3,381,400), we are left with $1,129,400, the amount of expenses that must be covered from savings and investments (from FC). We can call this amount ($1,129,400) the income floor, the minimum amount of FC needed to cover expected expenses. Note that the cushion of $695,200 from the balance sheet is the amount of FC above the floor (FC of 1,824,700 minus floor of 1,129,400 = 695,200 cushion). This represents, after holding back some amount for reserves, the risk capacity indicated by the balance sheet, or more simply, the upside. In this case, after holding back a reserve of $125,100 from the cushion of $695,200, the resulting upside is $570,100.

The final step is to lay this out as an allocation of financial capital, based solely on the strength of the household balance sheet (below).


Using the amounts above, the balance sheet shows an Upside/Floor/Longevity/Reserves allocation of 31.2%/61.9%/0%/6.9% (Longevity is a discussion for another day, but it generally starts with funding the deferral of SS benefits as the best deal around). Note that although this shows how much of FC can be exposed to upside risk and how much should be managed as floor, it is agnostic about how that should be done. In fact, as shown above under the “Adjusted” allocation, once you know what your balance sheet says, you can make an informed decision to expose more floor or less upside to risk, as you determine works best for you. In this case, about 9% of the floor has been allocated to upside (increasing the balance sheet upside allocation from 31.2% to an adjusted 40%), putting that much of the floor at risk—and therefore requiring careful management to prevent market losses from damaging the ability of the floor to cover expenses.

The Foundation for Solving the Puzzle

There is a misapprehension that using balance sheet risk management to allocate financial capital is somehow insurance product centric, puts safety-first above all, or is just liability matching in a different costume. It can be used to allocate any of those things, just as it can also be used to allocate a total return portfolio, a dedicated floor with upside, a bucket system, or a combination of upside portfolio, bond ladder, and insured products. All of those things are about implementation, and are independent of the mathematical determination of how much household FC should be managed as upside, floor, longevity, and held for reserve.

Balance sheet analysis comes before implementation. It replaces the narrow view of the investment portfolio and total-return allocation theories as the central focus of retirement with a broader understanding of the overall household financial situation and managing all the risks that can affect the retirement plan, not just investment risk.

The household balance sheet, not portfolio theory, is the foundation of personal financial management, anywhere in the lifecycle. A solid understanding of the household balance sheet provides the basis for a reasonable and practical way to solve the puzzle of how to best use household resources to fund retirement or reach other goals.

This is a brief introduction to a subject with a deep body of knowledge that is typically not part of an investment advisor’s agenda. For more information, check the reading list at the Retirement Income Industry Association website for links to additional readings and sources.

--Michael Lonier, RMA®






20 comments:

  1. I agree with the basic premise that some methodology for defining and determining an income floor is the most useful way to begin the task of coming up with a client specific allocation. However, I’m not clear what the advantages are of Mike’s more abstract and complicated balance sheet approach, which requires coming up with discounted present values for a lifetime of income and expenses vs. a simpler and more concrete back-of-the-envelopment calculation that subtracts current non-investment income from estimated current expenses, resulting in a net current dollar amount that represents how much income must be generated from investible assets, with the understanding that this amount will need to be inflation-adjusted.

    This is not a criticism, but rather a real question. For it seems to me the the second approach, which is how I’ve gone about determining my income floor need, does the trick sufficiently well and maybe even better given all the inherent uncertainties regarding future interest rates and stock market returns that drive the determination of a discounted present value, as well as the uncertainties of future income needs.

    ReplyDelete
    Replies
    1. The income/expense cash flow method you mention is fine as far as it goes as a first order estimate. The PVs that are used on the balance sheet come from cash flows over the plan horizon—SS, pension, earnings, other income, fixed expenses, one-time and irregular expenses, taxes, and so on. The more accurately you can model those cash flows, the less likely you will be surprised by something that breaks the plan. Once you have the cash flows, there’s no reason not to create the PV balance sheet. My tool does it all in about ten minutes if I have the client data in hand; after that, changes are instantaneous. Getting the expense detail is usually the longest lead item, so a summary top-down method for deriving that can be used as well as a first order estimate (current expenses = income minus savings). Most people know how much they make, and it’s pretty easy to figure out how much they save.

      Delete
  2. Barry, I'll ask Mike to respond, but if I understand your question correctly, the answer is your term, "current". You are assuming that all of your future liabilities and income are consistent with your current expenses and income. In reality, your future expenses and income will change as you add Social Security benefits, become less active, lose a spouse, have medical needs, etc. Discounting those expected changes provides the most accurate information.

    If your inflation-adjusted income and expenses throughout your retirement will never change from what they are today, then your approach and his would yield the same result.

    ReplyDelete
  3. (i) What about the value of your house? Is it ignored, or is it part of "financial assets"?

    (ii) The trouble with all DCF calculations is that the discount rate tends to be plucked out of thin air. How should its value be established for this purpose?

    ReplyDelete
    Replies
    1. Oh, my! Well, one should never pluck discount rates out of thin air.

      The purpose of discounting is to calculate the value of something in the future in today's dollars. You should discount a future investment value using the expected rate of return for that investment. If you are investing in stocks, you might choose a discount rate of 5% after inflation, for example, but if you were investing for that goal with bonds, you would choose an expected rate of return for bonds, which would be lower.

      If you need $1,000 twenty years from now and invest in stocks and expect a return of 5%, you would need $377 today. Invested in bonds and expecting 2.5%, you would need $610 today. You would use those expected returns as discount rates depending on which way you planned to invest.

      I would discount future expenses by the rate of expected inflation, remembering that not everything increases at the general rate of inflation. For example, education and medical costs deserve a more negative growth rate than the perhaps -3% you might choose for general inflation.

      I talked about houses in Three Portfolios. It is a financial asset, but the question is whether or not it is a retirement income asset. Unless you plan to sell it at some point, in which case it becomes cash with is a retirement income asset, or to take out a reverse mortgage, your home is a financial asset but not (yet) a retirement income asset.

      If you never plan to sell it or take out a reverse mortgage or rent it to someone, it will never provide retirement income, but it still a financial asset.

      Thanks for writing!

      Delete
    2. Thanks for replying. "You should discount a future investment value using the expected rate of return for that investment": if I knew that life would be much easier. When I worked for a large corporation, people pretended to believe that the discount rate we used was selected by the Chairman's wife. At least, I hope they were pretending.

      Delete
    3. I think she may have provided ours, as well.

      But you raise a serious point. We're making our best guess at what should be a conservative estimate of expected returns. If we guess a real return for stocks as 5% and use that for our discount rate and they do better, our plan will be fine. If they do worse, our plan won't meet expectations.

      The goal would be to show that your plan works with conservative discount rate estimates, but as you note, there is no guarantee than you will achieve even a conservative estimate. Still, it's far better than no plan at all.

      BTW, I suspect corporations choose discount rates that make their plans look achievable and the outlook rosy!

      Delete
    4. Thanks again. One more serious point: it's one thing to guess an expected return; what about the volatility of that return? Shouldn't that somehow affect the calculations?

      Delete
    5. It absolutely should.

      There are at least a couple of ways to deal with volatility. One is to use software that uses Monte Carlo simulation to predict the probability of outcomes based on the variance of returns. Another is to use an approach that I wrote about in Half Right. It's based on a Wade Pfau post that can be linked from mine.

      Delete
    6. I helped co-author a RIIA Update recently that specifies the discounting methods for planning. I’ll summarize here. When I get as link to the Update, I’ll also publish it here.

      First, the purpose of planning with the balance sheet is not to cut into stone an expectation of the personal wealth you will have at some future date so that when you meet up with yourself at the pre-appointed time, you will have x dollars still in the bank. No one knows how the future will change our lives. Rather, it’s a tool for assessing the relative value of your assets and liabilities so you can make a reasonable plan to meet achievable goals. The plan helps you do the things you need to do now to get where you want to go and provides a way of monitoring your progress as you make adjustments over the years.

      I like to run two plans with two sets of discounts—one long-term and one a current 1-year trailing average, to set up the guardrails on either side of the road. The data for all is available on public websites and the rates are easily derived (the details are in the RIIA Update).

      The long-term rate set is 3.32% for inflation (the 100-year average sine 1913 when the BOL started the CPI-U series, from bis.gov); 4.93% for discounting (the Schiller 100-year spliced long bond (from ’13)/10 year note (from ’53) interest rate, from Schiller’s website); and for expected rates of return, I use the 5-yr trailing average return of the assets in the plan from Jeff Considine’s QPP, which uses some autocorrelation and allows adjusting the S&P baseline up or down, with Harold Evansky’s CAGR adjustment to the arithmetic average, which uses the expected SD to reduce the mean. Human capital is discounted high, medium, and low based on stock, blended, and risk-free rates and a judgment of the client’s employment stability.

      I recalc the current rate set quarterly based on the 1-yr trailing daily average. Inflation as of 6/30/14 is 2.18% (from bis.gov) and the discount rate from the 10-year T-Note is 2.66% (from treasury.gov), and expected returns are handled as above.

      If you have a robust model, the different rate sets provide interesting results to study, and you begin to appreciate the effect of spreads between inflation, bonds, and stocks on a plan over time. Which hopefully gives you a better appreciation of using risk-free assets in your floor!

      Delete
    7. I helped co-author a RIIA Update recently that specifies the discounting methods for planning. I’ll summarize here. When I get as link to the Update, I’ll also publish it here.

      First, the purpose of planning with the balance sheet is not to cut into stone an expectation of the personal wealth you will have at some future date so that when you meet up with yourself at the pre-appointed time, you will have x dollars still in the bank. No one knows how the future will change our lives. Rather, it’s a tool for assessing the relative value of your assets and liabilities so you can make a reasonable plan to meet achievable goals. The plan helps you do the things you need to do now to get where you want to go and provides a way of monitoring your progress as you make adjustments over the years.

      I like to run two plans with two sets of discounts—one long-term and one a current 1-year trailing average, to set up the guardrails on either side of the road. The data for all is available on public websites and the rates are easily derived (the details are in the RIIA Update).

      The long-term rate set is 3.32% for inflation for escalating expenses (the 100-year average sine 1913 when the BOL started the CPI-U series, from bis.gov); 4.93% for discounting (the Schiller 100-year spliced long bond (from ’13)/10 year note (from ’53) interest rate, from Schiller’s website); and for expected rates of return, I use the 5-yr trailing average return of the assets in the plan from Jeff Considine’s QPP, which uses some autocorrelation and allows adjusting the S&P baseline up or down, with Harold Evansky’s CAGR adjustment to the arithmetic average, which uses the expected SD to reduce the mean. Human capital is discounted high, medium, and low based on stock, blended, and risk-free rates and a judgment of the client’s employment stability.

      I recalc the current rate set quarterly based on the 1-yr trailing daily average. Inflation as of 6/30/14 is 2.18% (from bis.gov) and the discount rate from the 10-year T-Note is 2.66% (from treasury.gov), and expected returns are handled as above.

      If you have a robust model, the different rate sets provide interesting results to study, and you begin to appreciate the effect of spreads between inflation, bonds, and stocks on a plan over time. Which hopefully gives you a better appreciation of using risk-free assets in your floor!

      Delete
  4. Hi Dirk. An interesting big picture approach. Michael posted a blog today as well on this ... a little different perspective on Morningstar's David Blanchett's recent research into this area ... http://www.kitces.com/blog/allocating-a-portfolio-in-the-context-of-human-capital-and-the-rest-of-a-households-total-balance-sheet/

    Interesting ... I commented on Michael's post that basically all the components of the balance sheet are likely to be a bit more volatile than our brains tend to imagine or project. As I mentioned during our recent conversation all data is stochastic!

    ReplyDelete
  5. I think there is a common thread in that all of a household's assets need to be considered. They might not all generate retirement income, but they shouldn't be ignored in any case.

    As I think I already pointed out on here somewhere, it's important for retirees to understand that their human capital portfolio is largely depleted by age 65. It's a more important consideration for savers, especially young ones.

    ReplyDelete
  6. This is one of the reasons why people dislike financial advisors. There's a lot of jargon here that says very little. The basic idea here seems to be that you should add up your expenses, and subtract those expenses from the PV of your SS + pensions, current savings, and PV of estimated future earnings. This all assumes a very simple and unrealistic model (constant inflation rate, capital return equal to inflation rate). No mention at all of portfolio allocation (and what it implies in terms of return and variance and sequence of returns risk), or whether "floor" assets should be invested differently than "reserve" or "upside" assets. Am I missing something?

    ReplyDelete
    Replies
    1. Joe, I believe what you are missing is the paragraph above that says:

      "Your savings and investments are just one part of the puzzle, one part of the resources you need to manage to get the best outcome. You need to start at a place that accounts for all of your resources, that balances what you have and what you need, revealing the weaknesses—the risk exposures—you need to overcome. Financially, the place to start is with your household balance sheet."

      Mike is saying that the household balance sheet is the place to organize your assets before you calculate all of those other things you say are missing, not that it replaces those steps.

      However, I am somewhat distressed, as you can imagine, to learn that people don't like financial advisors. That's a bummer.

      Thanks for writing, though!

      Delete
  7. Thanks for responding (and I was joking -- a bit -- about financial advisors)! I'm not arguing that it's not useful to estimate what your savings, future income, and future expenses will be. My complaint about this post (beyond too much jargon) was that this balance sheet approach really tells you very little, primarily because it implicitly assumes a certain kind of investment (returns equal to inflation), and there are no error bars associated with this model. Future returns on savings determine everything.

    Here's a concrete example. Suppose I take the 1.8M of savings used in this example and bury it in my back yard. Also suppose that inflation runs at 3% over my 30 year retirement. The PV of my FC is now about .7M, which means my .57M cushion is now -.53M. Not good -- time to cut back! Now suppose I invest my savings in a 60/40 stock/bond split and I'm lucky enough to get exactly 4% a year real returns for 30 years. Now my cushion is 4.6M and I'm already planning my round-the-world cruise. So what do I learn through the balance sheet approach?

    ReplyDelete
    Replies
    1. First and foremost, the backyard thing? Not the best approach.

      The balance sheet tells you very little about how you should invest, but that isn't its purpose. It allows you to organize and consider your assets. The planning you're talking about comes later.

      The balance sheet lets you identify how much your Social Security benefits are worth, for example, and how much you might be able to spend from savings. Will you sell your house and downsize? That needs to be accounted for (or not). Once you have an idea of the size of the income problem and the assets you have to solve it, the serious planning gets underway. A balance sheet is a good way to start.

      In your example, the backyard is probably not the way to go. Assuming you can earn 4% a year every year is the other extreme, and not a serious possibility. The answer lies somewhere in between and depends on the size of your shortfall and how much risk you're willing to take to cover it. Your balance sheet may indicate that you're in great shape, or that it isn't possible to reach your desired spending level. It is the starting point.

      As for the "jargon", I try to use as little as possible in my posts and to define it when I can't avoid using it. But retirement planning is sometimes difficult to explain without financial terms. I suspect there are only two options: learn the language and the process, or avoid the do-it-yourself route.

      Thanks for writing.

      Delete
  8. Very interesting, this whole balance sheet idea. I ran the numbers and am happily surprised to see that I appear to be a little bit overfunded! That does allay some of my new-retirement concern.
    Thanks for the good discussion.

    ReplyDelete
  9. Great discussion. Very intrigued by the balance sheet concept. I just finished a book called "Build Wealth and Spend It All" by author Dr. Stanley Riggs. I want to recommend this book as I finished it truly feeling like I have learned invaluable lessons about the future of my families finances. Most people are so hung up on retirement and how much money they will end up with, they don't take stock of what to do with it along the way. Dr. Riggs offers very down to earth advice and planning for the everyday person to not "get rich quick" but instead to build wealth...which is much more valuable in the long run. His website if you are interested http://buildwealthandspenditall.com

    ReplyDelete
    Replies
    1. Thanks, Dave.

      This raises an interesting point which is at the heart of a paper by Scott, Sharpe and Watson entitled, "A 4% Rule - At What Price?". The authors note that a constant dollar spending rule has the potential to build large terminal portfolios. That sounds good, but it also means that the retiree will spend less than he could have had he not left so much to his estate.

      Not every retiree wants to "spend it all", but not retiree wants to save it all, either.

      Delete