NN IP: A closer look at the US economy - Part 1

The booming US economy

The US economy is growing solidly above potential. At the most fundamental level, growth is supported by a strong feedback loop between income and spending in both the household and corporate sectors. The six-month trend in monthly payroll growth has increased from around 175K in the beginning of the year to around 220K in July. This is well above any reasonable estimate of break-even payroll growth (i.e. employment growth consistent with a stable unemployment rate in the long run). On top of this US corporates maintain their appetite for capex spending. This is evident in the fact that investment intentions in various regional and national surveys are holding up at historically high levels (even though they have come off the boil a bit). What’s more, despite some volatility, the uptrend in core capital goods orders and shipments remains pretty solid. Hence, so far the spill-over from trade worries to investment decisions appears to remain limited. On the consumer side we observed a strong bounce back in consumer spending from the Q1 weakness and there is every reason to believe this strength will be carried forward as well. Employment growth remains strong and wage growth is accelerating moderately. Consumer confidence remains very robust with the Conference Board measure reaching the highest level since 2000. 

Last but not least, the US consumer’s balance sheet looks very healthy. Net worth rose to an all-time high in Q1, mostly on the back of considerable financial asset and housing capital gains. Meanwhile, household debt as a percentage of disposable income remains around its 2002 level. When looking at the flow implications of this stock of debt, things look even better with the debt service ratio around the lowest level since the early 1980s. Of course some 35 years ago the debt stock was much lower, which explains the low debt service ratio back then. This time the main driver is the low level of interest rates. The good news here is that consumers have been shifting massively towards fixed rates over the past 10 years in an effort to lock in the historically very low rates. Rising rates will thus only filter though gradually into debt service ratios. Insofar as rising rates go hand in hand with better disposable income growth, the latter will act as a damper on the rise in debt service ratios. 

A final interesting observation about the balance sheet of the consumer is that the savings rate was significantly revised upward from around 3% to around 7% in Q2 on the back of an increase in the estimated level of consumer income, mostly from assets and property. The good news here is that there is more room for the savings rate to fall going forward, which could be a support for consumption growth once the tax cut effect fades away. 

To be sure, a necessary condition for this “savings rate engine” to provide more fuel to consumption growth is that there be a combination of solid consumer confidence and no significant declines in net wealth. The question is whether or not this condition is also sufficient. The upward revision of the savings rate opens up a huge gap between the actual savings rate and the savings rate one would expect historically on the basis of net wealth, credit standards and consumer confidence. Put another way, the marginal propensity to consume out of wealth has declined relative to its pre-crisis value.

There could be a number of reasons for this. For instance, the experience of significant wealth declines in 2008/09 could have made consumers more cautious in their willingness to spend out of wealth. Also, the measured aggregate marginal propensity to spend out of wealth may have declined on the back of rising wealth inequality as more wealth is now concentrated in high income/wealth brackets. The extent to which the savings rate will decline thus remains an open question. However, we can say that it acts as a potential upside risk for consumption in the foreseeable future. What’s more, if this risk does not materialize and the savings rate remains elevated relative to its fundamentals this will have positive long-term implications as well. After all, at some point net wealth is bound to decline again, in which case the drag on consumption will be less than it was in the past. That is unless the marginal propensity to spend out of wealth is bigger in the case of wealth declines than it is in the case of wealth gains. 

Of course the feedback loop between private sector income and spending receives considerable support from, and is augmented by, the fiscal boost. There seems to be a perception among some that this sugar high should already soon reach its peak. Although it is impossible to precisely pin down the peak because it depends on consumer and government spending behaviour, it stands to reason that the fiscal boost will only start to fade slowly sometime next year. We never subscribed to the view that the tax cuts were the major part of the fiscal boost for the consumer anyway, because they mainly impact high income households. The effect of the increase in discretionary federal spending caps on growth is much bigger. While these caps were raised in one go early this year, the actual spending that follows from them takes time to get going and its flow will thus last for a while. A non-negligible part of this spending will find its way through a boost in state and local government spending. In addition to this, the booming economy also contributes to more spending by sub-federal levels of government. State and local governments are required to balance their budgets, so their behaviour is usually relatively pro-cyclical.

All in all, the US economy thus seems on course to continue to grow well above 2% on a trend basis for the next four quarters. In the second half of next year and going into 2020 growth could then slow down to a below-trend pace as the fiscal boost turns into a drag and the effect of Fed tightening may be felt more strongly. The extent to which growth will slow down on the back of these factors is far from certain of course. The household savings rate may well be on a declining trend by then and help keep the economy going. Also, a marked pick up in productivity growth which is widely recognized by consumers and businesses could also give an important offsetting boost to both spending growth and supply-side growth.

Should we look at inflation in the traditional way?

This brings us to the next big debate in the US economy which is the extent to which wage and price inflation will improve further and possibly overshoot the Fed’s 2% target for core PCE inflation. There seem to be roughly two schools of thought here, both inside and outside the FOMC. The traditional school is mostly, but not exclusively, made up of academic economists and retains a lot of faith in the Phillips curve. As such this school emphasizes the existence of considerable time lags between labour market slack (i.e. the deviation of the unemployment rate from the infamous NAIRU or equilibrium unemployment rate) and wage and price inflation. While it is recognized that the NAIRU is surrounded by uncertainty, the traditional school assumes that the NAIRU only changes slowly over time and thus remains a useful concept for policymaking. There is even a possibility that the Phillips curve will become steeper when the unemployment rate falls well below the NAIRU. If this is the case inflation will unexpectedly accelerate, and the Fed may well feel the need to step up the pace of hiking. As a result, financial conditions could tighten substantially (think of the 1994 bond market carnage or the 2013 taper tantrum) which could potentially kill the expansion. Even if the Phillips curve does not contain a non-linearity, overheating may still be a dangerous strategy because all kinds of (financial) imbalances could grow larger in this state of the economy. This could also shorten the remaining lifetime of the expansion. Even the existence of uncertainty about whether the Phillips curve will become non-linear should be enough to dampen the incentive to overheat. After all, once inflation accelerates and threatens to pull inflation expectations up the Fed will need to induce a rise in unemployment to get it down again. In that case the economy will re-enter the “flat” part of the Phillips curve so the required rise in the unemployment rate may be quite substantial indeed. This is essentially the labour market flipside of the idea that the distribution of economic outcomes has a larger downside skew in the vicinity of full employment.

The non-traditional school recognizes all these arguments but emphasizes the fact that the NAIRU and the economy’s potential level of output are subject to considerable uncertainty even in the best of times. What’s more, over the past few years these concepts may well have been subject to structural changes which work in a disinflationary direction, either because they push the NAIRU lower or because they reduce the inflationary consequences of a given degree of overheating. These structural changes have taken place in both product and labour markets and relate to the effect of more online competition and pricing transparency, higher levels of industry concentration, lower worker bargaining power etc. Combined with the fact that inflation has been below the Fed’s target for many years while inflation expectations are well-anchored, these considerations are sufficient for the non-traditionalists to probe the extent to which monetary policy can push the economy on a permanently higher growth path. Some readers will already have noted that Greenspan was very much a non-traditionalist in his time. In the second half of the 1990s the US unemployment rate fell below real-time Fed estimates of the NAIRU, but Greenspan correctly assumed that it could fall further without generating more inflation. The reason is that Greenspan spotted an increase in the underlying rate of productivity growth early on. 

The future trajectory of productivity growth indeed plays a crucial role in the debate between traditionalists and non-traditionalists. Higher productivity growth will allow decent real wage and real profit growth to coexist more peacefully. In a way it will cause the pie itself to grow faster which dampens the incentive for each of these actors to try and grab a bigger share at the expense of the other. Or, from a different perspective, higher productivity growth slows down the pace of decline in the unemployment rate and pushes the NAIRU lower. The latter stems from the fact that higher productivity growth raises the expected surplus from posting a vacancy, which will lead to more vacancies per unemployed person unless workers succeed in capturing the entire additional surplus. In addition to productivity growth, structural changes in worker bargaining power also feature prominently in the debate about the risks of probing the economy’s supply side. We will delve into that somewhat more next week and discuss the implications for the Fed.

Emerging markets: widening growth divergence

Despite the intensifying market pressure on emerging markets and the dramatic news headlines coming from some of the problem countries, the EM growth picture remains reasonable – if we look at the aggregate numbers, that is. Once we zoom in and look at the individual countries, we see rapidly growing differences. 

Let’s start with the overall growth picture. The weighted average EM GDP growth is slowing from the Q1’18 peak, but only modestly. Our forecast for 2019 is 4.8%, compared with 5.2% this year. Growth in 2017 was also 5.2%. But behind these numbers we see a widening divergence between the countries with limited macro imbalances and an effective economic policy mix versus those with large imbalances and authorities that are unable or unwilling to adjust policies to the new challenging environment.

Most of emerging Asia is in relatively good shape. There is the threat of an escalating trade conflict between China and the US, which could have a sizeable impact on non-Chinese exporters in the region, but the domestic economies should be able to offset some of this potential negative effect. China itself has started a new round of economic stimulus, which should more than offset the negative impact from higher US tariffs. Also, in the past week new measures were announced, including a lowering of income taxes. India, the region’s second-biggest economy, has limited vulnerability to global trade problems because its economy is relatively closed. Moreover, the country has started a new credit cycle after years of deleveraging. This should drive an acceleration in both consumption and fixed investment growth in the coming quarters.

Outside Asia, there are other countries with decent macro fundamentals that will not necessarily experience much of a slowdown. Chile is a good example, but even Mexico looks reasonably well-positioned to keep its growth rate stable after the bilateral trade deal with the US. Much will depend on the economic policy course of the new president, Andrés Manuel López Obrador, who will be installed on 1 December. Russia is benefiting from the rising oil price and should be able to avoid a slump. New US sanctions are a risk, but oil should remain the dominant factor.

The growth problems are concentrated in the fundamentally challenged countries, where the recent market turmoil has been the most painful. Turkey continues to resist a major tightening of economic policies, but it is just a matter of time before interest rates will be hiked substantially and credit growth will quickly decline. It is highly unlikely that Turkey will be able to avoid a deep recession. The economy in Argentina has already started to contract after the government was forced to tighten policies in a bid to reduce the country’s external financing needs. More adjustment will be necessary, and the recession is likely to deepen in the coming quarters.

Of the three second-tier problem countries – South Africa, Indonesia and Brazil – the first has already fallen into a new recession. The sharp weakening of the rand and the likely policy response of the central bank can only further dampen the economy. Still, we do not expect a dramatic adjustment, as the government is generally seen as reformist and the central bank as credible. Besides, the country’s macro imbalances have been reduced substantially in the past years and the banking system is in good shape. 

Indonesia can still avoid a sharp slowdown if its government tightens its grip on the economy and resists the temptation to pursue more interventionist policies ahead of next year’s elections. Earlier this week, the government announced that it will delay half its ambitious electricity projects in a bid to reduce imports and stop the widening of the current account deficit. This will have a negative impact on growth, but steps like this might reduce pressure on the rupiah for the short term. Indonesia remains one of the most interesting countries to watch in the current EM market stress. The authorities continue to intervene heavily in the foreign exchange market, which is keeping the rupiah stronger than it should be. In the current environment, with high pressure on EM currencies and a rising oil price, the rupiah looks vulnerable.

And finally, Brazil: here the verdict is still out. Economic growth has been remarkably resilient in the past quarters. The recent sharp drop in the real has increased the likelihood of interest rate hikes, but it is important to remember that the country does not have a large external imbalance. The big policy adjustment has to be on the fiscal side and the October elections will give clarity on this. Chances of a reformist candidate winning look slim, at least for now. Uncertainty therefore remains high, which is less than ideal in the current market environment.