It’s a never-ending ritual: many people hype current (and try to guess upcoming) national income statistics such as Gross Domestic Product – or heed those who do. The bigger is GDP and the more rapidly it rises, mainstream economists and policymakers strongly imply (and investors and journalists obediently accept), the healthier is the economy and the better are investors’ prospective returns.
This practice is pointless and myopic. It’s futile because there’s no clear or consistent relationship between GDP’s rate and direction of change and stock markets’ returns (see, for example, Three risks you can discount – and one you can’t). And in this article I demonstrate that the obsession with macro-level data ignores two crucial micro-level facts:
- Many American households’ CPI-adjusted, pre-tax income was little greater in 2019 than it was in 1989 – and in some instances has fallen since the GFC.
- Most households’ wealth has hardly risen since 1989, remains at a low level and has sagged since 2007.
Over the past 30 years, America’s “real” GDP has grown almost 2.5% per year; net of CPI, in other words, it’s doubled. Yet only a minority of its households is now richer – and many are poorer – than they were then. In effect, and regardless of GDP’s size and rate of growth, during these three decades many households have endured a seemingly-permanent state of recession. That’s hardly the American Dream; it’s more an American Nightmare.
Is the situation nearly so dire in Australia? Probably not – yet – but we fool ourselves if we assume it’s greatly different. The lesson is obvious: investors should discount or ignore short-term changes of highly-aggregated national income statistics. These fluctuations tell you little of any relevance and obscure much of critical significance.
The Fed’s Survey of Consumer Finances
This article summarises my analysis of the Federal Reserve Board’s triennial Survey of Consumer Finances (and secondarily of data compiled by the Australian Bureau of Statistics). SCF provides the most detailed, valid and reliable at a given point in time, comparable across the greatest extent of time and publicly-available estimates of American households’ finances. Every three years since 1989, it has provided detailed information about incomes, assets and liabilities. The Fed conducted its most recent SCF in 2019; the next one is now underway, and its data – which will quantify the effects of the COVID-19 lockdown and recession – will be released in 2023-2024. In my more idealistic younger years, I was astonished that the Fed and its acolytes in the mainstream media resolutely disregarded these data. In my more jaded middle age, I understand why: they tell an inconvenient truth. (My only remaining surprise is that the Fed continues to conduct the SCF.)
Importantly, the Fed’s and ABS’s surveys don’t repeatedly survey the same people; instead, they draw a new sample of different individuals every three years. To cite one of many examples: people aged 45-54 years in 1989 and 1992 aren’t quite the same people; some who qualified in 1989 were too old in 1992, and some who were too young to 1989 qualified in 1992, etc. We must therefore draw inferences carefully.
American Households’ “Top Line” 1989-2019
Figure 1 plots American households’ median pre-tax income. (One-half of households earn more than the median, and one-half less). It disaggregates income according to the age of the “reference person” whom the Fed contacted to complete the SCF. The categories of age are:
- 45-54 years (most adults’ prime income-earning years),
- 55-64 (people nearing or possibly in retirement) and
- 65-plus (probably retired).
These and the following graphics remove the effect of “inflation” (as defined and measured by the Consumer Price Index); they thereby provide a long-term “apples-to-apples” comparison of the purchasing power of households’ income during this 30-year interval.
Figure 1: Median Household Pre-Tax Family Income, Stratified by Age, Thousands of Constant (2019) $US, 1989-2019
Four points are paramount:
- since 1989, most American households’ pre-tax income has barely grown – and for some, it’s decreased;
- the GFC was a watershed: before it, most incomes tended to rise; since then, most have stagnated and many have fallen;
- if anything, the rot preceded the GFC; in other words, the GFC accelerated rather than caused the trouble: in households as a whole, as well as those whose reference person was aged 45-54 and 55-64, pre-tax income fell continuously from 2004 to 2013;
- overall, household incomes haven’t just risen glacially; they’ve increased much more slowly than GDP (also measured in constant (2019) $US, Table 1).
The median American household’s pre-tax median income was $51,900 in 1989, $58,500 in 2004 and $59,100 in 2019. That’s a compound annual growth rate (CAGR) of 0.43% per year over 30 years. Before the GFC, the median household’s income rose 0.66% per year; since 2007, it’s risen by just one-tenth of 1% (0.10%) per year. Households whose reference person is aged 55-64 years, as well as Hispanics and holders of tertiary degrees, have suffered most since the GFC.
Table 1: CAGRs, Household Pre-Tax Income, Stratified by Age, Schooling and Race
The 65-plus group is the only one which has fared continuously better over 30 years and escaped the GFC’s ravages: in these households, median pre-tax income has risen almost 1.5% year since 1989, and slightly faster since the GFC than before it. As a result, in 1989 their income ($30,100) was 58% of the median; in 2016 ($64,600) it was 85% (80% in 2019).
Table 1, which stratifies households along additional criteria (it omits others, such as family structure, etc., to which these same points apply), substantiates Figure 1. In virtually no case has household income risen as fast as GDP – and in most cases, it has greatly lagged GDP. It’s much better if the household’s reference person has a tertiary degree (median household pre-tax income of $95,700 in 2019) than just a high school diploma ($45,800) or no secondary certificate at all ($30,500). Yet since 1989 “tertiary” households’ incomes have grown even more slowly than those without any qualification; they’ve also fallen relatively rapidly since the GFC. Similarly, white households’ incomes are much higher ($69,200 in 2019) than blacks’ or Hispanics’ (both ca. $40,700); yet whites’ households’ incomes have grown much more slowly than minority households’.
Figure 2: Median Household Pre-Tax Family Income, by Quintile, Thousands of Constant (2019) $US, 1989-2019
Over the past 30 years, American households’ incomes have mostly stagnated – and the incomes of more than a few have fallen. I’ve already noted that households whose reference person is aged 65-plus break this general rule. Similarly, although they’ve not grown as rapidly as “real” GDP, the pre-tax incomes of high-income households – those in the top quintile (20%) of income-earners – have grown much more rapidly than those of average- and low-income households (see Figure 2, which omits the lowest 20% of households because their pre-tax income is so low – $12,500 in 1989 and $16,300 in 2019 – that it’s barely visible).
In 1989, the median income of the top (#5) quintile was $163,000; in 2019, it was $221,000. That’s a CAGR of slightly more than 1% per year – less than the growth of “real” GDP, but much greater than the other quintiles. Further, in this top quintile the rate of growth during and after the GFC was almost as high as it was before the crisis; in the lower quintiles, post-GFC rates of growth have mostly collapsed (Table 2).
The numbers in Quintile #2 of Figure 2 are particularly sobering: in 2019, the median income of the distribution’s 20th-40th percentile was $35,600. In other words, half of this quintile (comprising 10% of all households) and all of Quintile #1 (20% of all households) – that’s almost one-third of all households – earned pre-tax incomes of $35,600 or less per year. Recent research corroborates this result: a month ago, Oxfam America published a study entitled The Crisis of Low Wages in the US. It found that 32% of that country’s workforce – almost 52 million workers – earns less than $31,200 per year.
One-third or more of American households, in other words, comprise the “working poor.” The soaring prices of essentials – food, petrol and housing – have surely made it even more difficult for them to make ends meet. For decades (details omitted for the sake of brevity), they’ve tried to cope by borrowing. As a result, the lower is the household’s income the higher is debt relative to income. But rates of interest are now rising – which will also intensify the pressure upon low-income households’ finances.
Figure 3: Median Household Pre-Tax Income, by Housing Status, Thousands of Constant (2019) $US, 1989-2019
Figure 3 disaggregates household pre-tax income according to housing status. Since 1989, renting households’ incomes have grown more quickly than homeowners’ (Table 2). Yet home-owning households earn 20-30% more than the median, and 2.0-2.5 times more than renting households. In 2019, the medium income of home-owning households was $77,400; in renting households, it was $35,600. In the U.S., home ownership has become the preserve of the better-off.