Investment Note: Death by debt? Not yet.

We keep hearing that there is too much debt in the world, yet the most synchronised hiking cycle on record has not yet resulted in a major slowdown in global economic activity. Even in South Africa, the highest repo rate since 2008 has not derailed consumer spending. What is going on?

The simple answer is that although interest rates and debt levels are clearly linked, there are other factors at play. It is not primarily the amount of debt that matters, but rather the ability to service that debt. For households, this means the ability to make monthly mortgage or vehicle loan payments. These payments usually combine interest and amortisation, in other words, the capital amount of the loan is spread out over the life of the loan. 

Where income comes in

So, while the interest rate on the loan is important, as is the size of the loan, income is equally important. When there is strong income growth, consumers can take increased monthly payments in their stride. Conversely, when income falls, even a small amount of debt at a low interest rate can become burdensome. 

For larger borrowers, like companies or governments, loans are often at a fixed interest rate and over a fixed term. These loans are often rolled over at maturity. For instance, if the government issues a 10-year bond, it pays a fixed interest rate for the life of the bond, and usually rolls it over maturity, meaning that, rather than repaying the capital amount (bonds are usually not amortised), a new bond is issued to the holder. This implies that the debt servicing ability is not only determined by the factors listed above, but also the timing of the maturity of the loan. If a given company or government’s bonds mostly only mature in the years ahead, the impact of recent rates hikes will clearly be blunted. If maturing loans must be rolled at much higher rates, it can be problematic. Even worse, if the loans or bonds mature in an environment of market stress – like in March 2020 – the borrower might struggle to roll them entirely and must suddenly find the cash to repay the capital. 

We’ll focus on household debt servicing today. However, the increase in government debt levels over the past few years is notable and a topic we’ll return to in the near future. 

Starting with the US, it is striking how little the debt service burden has increased for households. Despite the 500 basis points of interest rate hikes in this cycle, the most since 1980, the ratio of debt service costs to disposable income is hovering near record lows, as chart 1 shows. 

Chart 1: US household debt service ratio and interest rates

Source: Refinitiv Datastream 

We care about the US consumer since it is the big engine of the global economy. US annual consumer spending is equivalent to the dollar amount of total Chinese economic activity. At this particular point in time, we also care about US consumer spending as it is unlikely that we’ll get a meaningful slowdown in inflation while consumer demand is strong. Put differently, we care about the US consumer since the Fed cares deeply. 

Why has US consumer spending not been crushed by significantly higher interest rates? Simple. Debt levels are still relatively low as households repaired their balance sheets in the wake of the 2008 Global Financial Crisis, a Made in America crisis that was the result of too much household debt, and too much financial engineering by banks. The ratio of household debt to US GDP declined from 100% in 2008 to 74% today. Meanwhile, nominal income growth has been very strong (debt is usually expressed in nominal terms, and high inflation reduces debt burdens. On the other hand, deflation makes debt worse, as we’ll discuss in the case of China below). 

The US labour market remains tight. Unemployment was 3.6% in June, near record lows. Payrolls grew at an annual pace of 2.5% in June, slower than recent months but still well above the 20-year average of 1%. This is translating into wage growth of 4.5% to 6%, depending on the measure, a healthy pace of income growth by historic standards. 

Chart 2: Household debt to GDP, %

Source: Bank for International Settlements 

Finally, US households have the major advantage of largely taking out fixed-rate mortgages. Unlike in South Africa, where mortgage rates adjust automatically with changes in the repo rate, most US mortgage rates don’t change during the lifetime of the loan. 

These reasons explain why higher interest rates had limited traction in slowing consumer spending. It suggests that rates would have to rise further and stay elevated for longer to move the needle. 

Wealth in good health

Another factor to remember is that debt is only one side of household balance sheets. If the asset side is healthy, it does provide a cushion. In a pinch, households can sell some of their assets to meet cashflow problems. A worst-case scenario is high debt levels and falling asset values, exactly what we saw in 2008. US house prices are off their peaks, as are equity prices, but overall household net wealth is in good health (though it is concentrated in the hands of higher income groups). 

What about elsewhere? Countries like New Zealand, Canada and Australia seem particularly vulnerable due to the combination of sharp increases in household debt in recent years and a higher portion of floating-rate mortgages. However, with inflation still running hot, all three central banks are still in hiking mode.   

Opposite direction

Then there is the case of China, where rates have been falling of late, moving in the opposite direction to America and other developed countries. Chinese consumer debt as a share of GDP is not high by global standards. However, it has risen faster over the past decade-plus than any other major economy, tripling from 20% in 2009 to 60% today. As household income is an unusually low share of GDP in China, the ratio of debt to income is similar to developed economies. Again, this matters because it is income that needs to service the debt. 

It is true that Chinese consumers also have high levels of savings, meaning there is a cushion. The problem is that as in most other countries, mortgage debt is the largest component of overall Chinese consumer debt, and the asset linked to this debt – residential property – is teetering. House prices have been falling, which makes consumers feel very vulnerable and reluctant to go out and spend. China’s consumer confidence index plunged from a pre-Covid level of around 120 points to 85 points late last year. By March, it had only recovered to 94. It is still lower than it was at any point between the start of the survey in 1991 and the onset of the pandemic. 

Household debt in China is only a third of corporate debt, which is the real problem area. Much of the debt is linked to struggling property developers, and a fair portion is denominated in US dollars at a time when the renminbi has been falling. Local governments also have large debts and have until recently relied on selling land to developers as a key source of income to service the debt. This is going to be a struggle in future as demand for land dries up. 

Down, but not out

In South Africa, interest rates have also risen rapidly over the past 18 months. How vulnerable are local consumers? The question matters, since as much as we focus on the negative impact of loadshedding on the economy, consumer spending remains the largest component of economic activity. 

As is the case globally, we need to consider how debt levels have increased relative to growth in incomes. Incomes have grown faster than household borrowing for most of the past decade. Unlike Australia, Canada and New Zealand, the local property market has been lacklustre over the past decade, particularly outside the Western Cape. Real house prices have flat-lined since 2010 and are in fact still lower than the peak of the pre-2008 boom. Moreover, most South African households do not have a mortgage. While many lower income households might be overindebted with unsecured loans, these are small in comparison with home and car loans. One car loan can be equivalent to hundreds of in-store loans, for instance. 

Income growth has also been fairly resilient, apart from the brief slump in 2020. This is clearly an aggregate indicator, and always problematic in a very unequal society such as South Africa. But it is the aggregate that matters for overall economic growth. Nominal household disposable income grew 7.5% year-on-year in the first quarter, in line with the average of the past decade. Interestingly, wage income only grew 4%, so a large portion of recent income growth has come from other sources, such as investment and business income. 

Chart 3: South African household debt to income ratios, %

Source: SA Reserve Bank 

As a result, the portion of household income going to interest payments increased with the Reserve Bank’s rate hikes, from 6.4% in 2020 to 8% in the first quarter of 2023. This takes us back to the average of the decade before Covid. It is nowhere near the two historic extremes, 1998 when the prime rate surged to 25%, and 2008 when debt levels peaked. Both those spikes preceded a sharp slowdown in household spending. 

Gloomy

Consumers are gloomy, with the FNB/BER Consumer Confidence Index falling to -25 index points in the second quarter, one of the worst readings in the survey’s 40-year history. Notably, however, there was a record gap between respondents’ perceptions of the direction of the South African economy and perceptions of their own household finances. In other words, consumers were extremely bearish on the prospects for the economy over the next year, while only somewhat negative on their own financial position. As is the case with business and financial market sentiment, the pervasive pessimism about the country’s future seems excessive compared to the reality. 

Implications

Since we started with interest rates, we’ll finish with the monetary policy implications. In the US, the Federal Reserve’s job is not done. The slowdown in consumer spending needed to cool inflation remains absent, partly because consumers don’t yet feel the squeeze from higher interest rates. However, given that interest rates work with lag into the future, while the data at hand comes from the past (also with a short lag) there is always the risk of overtightening. While the analogy of landing a plane has been used a lot in recent months – i.e. a hard landing or soft landing for the economy – flying a plane is a real-time activity. The plane responds immediately as the pilot works the controls. Imagine landing a plane where you only know yesterday’s weather conditions, the instrument panel gives you readings from a few hours ago, and where your actions will only affect the plane later today. 

Some countries like Australia might already be in very vulnerable territory due to the build-up of household debt in recent years. Australia is not systemically important, however. China is, but it has a set of very different problems to the US. 

In South Africa, the good news is that consumers are possibly a bit more resilient than widely believed. Unlike the US, this does not mean the Reserve Bank has to hike rates further, since local inflation has largely been a supply side phenomenon, and not driven by strong household demand. However, the Reserve Bank will be influenced by what happens in the US, and if the Fed accelerates the pace of hikes, the Reserve Bank will be under pressure to stay in the game.