In these unusual times we now live in, worrying about if households, sovereignties and everyone in between have too much debt is certainly in vogue. One can understand why as, at least for in the U.S., it is a lynch-pin component to determining whether we have now entered the decline phase from Pax Americana.
If we are indeed saddled with too much debt, this financial crisis may well have accelerated us into a decline from which we may never truly recover. If not, we’re down presently, but not out. From a stock market point of view, if the former is the case, several years from now the Dow will be 5000 – and every long term investor who has been trained over the past several decades to buy dips will be colossally burned. If the latter is the case, it will be more like 20000 – and this moment in time will come to be regarded as one of the best in a generation to have bought stocks (and real estate, for that matter).
Right now, I want to focus on the U.S. household (I may address government debt in a subsequent article). In essence, it is the core of it all – households provide labor and revenue for business, as well as the majority of income for governments. So is the household, the core of our society, in too much debt?
Let’s first analyze the household as if it were an operating company. We may refer then to personal income, less consumption (which is before interest expense, taxes and other transfer payments) as EBITDA (earnings before interest, tax, depreciation and amortization expenses). We might use some common financial analysis metrics such as interest coverage (in this case EBITDA to interest expense), EBITDA margin (household EBITDA as a percent of personal income), debt to EBITDA, debt to equity, etc. These are indeed tools that the ratings agencies (for whatever that’s worth anymore) use when determining credit ratings.
So using those metrics, what does "U.S HouseCo, Inc." look like? Observe the charts below that I have built (source data, BEA and the Fed):
Reading through Moody’s various credit ratings methodologies, I might characterize each metric determining a rating as follows:
1) Prior to the early 1980s, household interest coverage of 11-14 times would have received a Aaa rating. Afterward, it would have slipped to Aa and presently – despite a current uptick due to unusually low interest rates – it might be flirting with an A at 8-10 times.
2) Prior to the early 2000s, household EBITDA margins of 19-25% would have received between Aaa to Aa, but lately below 19% might get an A.
3) Prior to the mid-1980s, household debt to EBITDA of 2.5-3.5 times had been at the cusp of investment grade at Baa. It might have dropped below investment grade in the 1990s and has presently collapsed to one of the worst junk ratings, Caa, after the mid-2000s at ~6.5 times.
4) Prior to the early 1980s, household debt to book capital (or equity) 15% or lower enjoyed a Aaa rating. Presently, having skyrocketed recently to 25-30%, it is now maybe an A.
Now obviously what I just did there is completely arbitrary on my part. But I believe that one can start to get a sense from this that the U.S. household has been in a state of decline in creditworthiness for decades - having accelerated in the past couple-few. I also observe that three of the four metrics possibly suggest that the household is still quite “investment grade”, whereas one indicates total junk status. One might conclude from this that the U.S. household is still in relatively good shape, but has eroded from sort of a Aa, to somewhere in between A and Baa rating. This is hardly junk status – but it something to keep an eye on for sure.
Those kinds of metrics have value when analyzing operating companies – entities that are set up to generate profits for its owners as a going concern in perpetuity. But households aren’t quite like that. They have a finite nature to them. Aside from getting the kids through college (and out of the house), as well as some other stuff, household wealth – their saved income accumulated over time – pretty much exist to serve as a source of income for an individual’s retirement years.
Here is how I look at the true relevancy of household debt; my sort of “final analysis”. Households have assets. The house, the car and all the stuff in the house are tangible assets. But those are generally illiquid and tend to have more of a utility than “investment” value. So I focus on the more liquid financial assets only – cash, stocks, bonds, retirement accounts, life insurance policies, etc. I look at that net of total household debt – after all, that “net financial wealth” is all that we will have, subsidized by Social Security benefits, to get on in our retirement years (presumably still 65 years or older). So how does the total retirement burden of the U.S. compare to the total of household net financial wealth (again, subsidized by SS)?
Firstly, how do I look at defining the total retirement burden? In 1950, people generally died of old age when they were 68. Today they die on average at 78. That means that, in 1950, on average they pretty much just had to worry about funding three years of retirement. Today it is 13. Back then, there were 12 million people older than 65 (~8% of population). Today there are 40 million (~13% of population). So in 1950, we needed to somehow fund on average 39 million retirement man-years (three years of life expectancy beyond 65, times 12 million people). Today we must fund 520 million retirement man-years.
In 1950, annual disposable income per capita was ~$1,400. So, in dollar terms, we needed to fund ~$55 billion worth of retirement man-years (~$1,400, times 39 million man-years). Today disposable income is ~$36,300 per capita. So if retirees wish to continue living the lifestyle they are accustomed to at that level of income (irrespective of inflation), we need to fund $18.9 trillion worth of retirement man-years. The latest Federal budget statements say we paid out nearly $700 billion in Social Security benefits last fiscal year. That works out to be ~$17,000 per person older than 65. If we adjusted disposable income for this subsidy, we still need to fund about $10 trillion of retirement burden.
This is where our nest egg – household net financial wealth – comes in. Presently, households (really inclusive of non-profit organizations) have $45.5 trillion of financial assets. They owe $13.9 trillion (mostly the mortgage). That means their net financial wealth is $31.6 trillion (all source, the Fed’s Flow of Funds report, through March 2010).
So, the sum of all U.S. household net financial wealth is presently a little over three times the total U.S. retirement burden net of subsidies, as I define it ($31.6 trillion, over $10 trillion). In 1950, it was 12 times the then burden. In other words, in 1950, we had 12 years’ worth of liquidity to fund three years of retirement (net of subsidies). Today, we have three years of liquidity to fund 13 years of retirement. We are completely upside down relative to decades ago.
And it only looks worse going forward. The Baby Boomers (the first of which born in 1946) are only now getting ready to start retiring months from now. According to Congressional estimates, those older than 65 will plateau in 2030 at 72 million, or ~20% of then population. Life expectancy will be just under 80 then. If they don’t raise the retirement age (and I have my doubts they won’t), we will need to fund one billion retirement man-years in 2030.
Let’s say we grow disposable income at 5.1%, Social Security at 5.5%, household financial assets at 6.0% and household liabilities at 7.2% - simply because those are the respective average annual growth rates for each in the past 20+ years (but no one is saying that past performance is indicative of future results). The result, using this “final analysis”, is household net financial wealth will only be able to fund 1.5 years of the then 15 required. Again, this is all net of Social Security subsidies, and assuming those are not somehow diminished, or otherwise compromised.
Therefore in 20 years one-fifth of the U.S. population will almost certainly be in some mode of significant belt tightening. And that cannot portend good things for the business of America – and thus the long-term prospects for U.S. stocks. Now I hate being a “nattering nabob of negativism” - and I do love my country - but someone please tell me how I get comfortable with how to successfully accommodate for this scenario?
An allegedly resultant higher savings rate improves the analysis, but pushing some numbers around, it seems too little, too late – especially considering the possible negative feedback loop on investor returns resulting from this scenario. Opportunities do exist – ranging from relaxing immigration laws (making the average age younger) to strengthening exports. But I just don’t see policy heading in that direction (at least not presently). And how do you get elected in this country running on a campaign of taking more and more, cutting benefits and telling people to live more modestly and work longer and harder for less?
All of that isn’t even the worst part of this. Here’s the final killer. Most of that household wealth is not in the hands of most households, whereas debt and the ultimate retirement burden is. The top 1% wealthiest households own one-third of all U.S. wealth. The top 20% own 85% of all wealth. In other words, 80% of Americans only own 15% of that $45.5 trillion needed to fund their retirement. So unless Bill Gates or Warren Buffet is going to pay off our mortgage and send us a check every month during our golden years (beyond that they already are through their income taxes), we have a problem. In my view, this problem serves to highlight the looming redistribution of wealth pressures coming down the pike.
So how much is too much or too little household debt? The most truthful answer I believe I, or anyone else for that matter, could give you is – who knows? But it sure feels like we’re a lot closer to too much than too little.
Disclosure: No positions.