How productivity growth affects GDP growth

If you have followed my writing for a while, you will recall that I come back to the following economic identity at regular intervals:

I. `Delta`GDP = `Delta`Workforce + `Delta`Productivity

Now, let me explain why that is the case. As I am sure you will agree, total output in an economy always equals the number of hours worked multiplied by the output per hour, or:

II. Total Output = Hours Worked x Output per Hour

Using simple maths, that formula can be expressed as follows:

III. `Delta`Total Output = `Delta`Hours Worked + `Delta`Output per Hour

Now, think of total output as GDP, and think of output per hour as productivity. All we need to prove now is that `Delta`workforce = `Delta`hours worked, and we will have proven that (I) equals (III).

There are no reliable macroeconomic statistics on the total number of hours worked so, at first glance, we are struggling a bit. Thankfully, `Delta`workforce, when looked upon in aggregate, is 99.9% correlated with `Delta`hours worked, and we have very reliable statistics for how the workforce changes from one year to the next; hence one can replace `Delta`hours worked with `Delta`workforce without doing any meaningful damage to the equation. In other words, (I) is not only similar to (III), but the two are almost identical.

One more point. Think of (I) as a reliable indication of trendline growth in the economy, but never use it to estimate GDP growth from one year to the next. Although annual changes to the workforce can be estimated quite precisely over shorter time periods, `Delta`productivity is a residual factor, which is only calculated ex-post, i.e. once we know what `Delta`GDP is. In other words, cyclical factors make `Delta`productivity jump up and down quite violently in the short term.

In practical terms, as you can see in Exhibit 1, the US economy has suffered a significant slowdown in both workforce and productivity growth more recently – and the same has happened in the rest of the world. Going forward, we know that workforce growth will be even more modest and, in some countries, even negative (see more on that below), but we don’t know what will happen to productivity.

‍

Why productivity growth is so important

All this is important for one very simple reason. We know that the workforce will decline in most countries between now and 2050, and we know that the worst affected countries will see an annual decline of about 1% (Exhibit 2). We therefore know that GDP growth will be dramatically negatively affected, unless we can find ways to boost productivity.

Negative GDP growth is not the end of the world, if the country in question is in good shape with low financial leverage (e.g. Germany and Switzerland), but that is not the case in many countries with major demographic issues (e.g. Italy and Japan).

‍

As you can see from Exhibit 2, within the OECD, the US economy is in best shape demographically with the US workforce expected to grow by about 0.25% annually between now and 2050. Having said that, strict immigration rules applied by the Trump regime may still affect that estimate negatively, and that could devastate entire industries - e.g. US agriculture, which is very dependent on Mexican labour.

Different measures of productivity

I should clarify how productivity is typically measured, because it is not as simple as you might think.

One measure of productivity is what is called labour productivity. It measures the output per hour worked, and is the measure of productivity that most focus on.

There is a problem with that measure, though. Labour productivity will rise sharply, if sufficient money is spent on new machines, but that doesn’t necessarily improve overall economic efficiency.

Consequently, the concept of total factor productivity (TFP) was conceived. It is calculated as the percentage increase in output that is not accounted for by changes in the volume of inputs of capital and labour. In other words, TFP is a measure of what share of increased productivity can be explained by factors other than growth in labour or capital.

Over the past half century, almost two-thirds of the growth in TFP can be explained by technology improvements. It is a better proxy for an economy’s return on capital, but it wouldn’t be fair to simply replace labour productivity with TFP. They are two very different measures of productivity.

Why is productivity slowing?

I often come across the argument that slow productivity growth since the crisis in 2008 is the key reason why GDP growth has been so pedestrian more recently, and it is indeed correct that productivity growth has been very modest since 2008 (Exhibit 3).

Having said that, it is a fallacy to think that slowing productivity growth only dates back to 2008. As you saw in Exhibit 1, US productivity growth began to slow in the late 1980s, and UK productivity growth has also been losing momentum since the 1980s. Germany’s picture is murkier, but that is most likely a function of the benefits that were harvested in the years after German reunification.

As a consequence of this observation, it is only reasonable to ask what could possibly be behind this phenomenon. If slowing productivity growth is not at all the post-crisis phenomenon nearly everyone claims it is, something else must be astray. What could that be?

Ageing of society at large is one reason why productivity growth has lost momentum for all these years. Whether we like it or not, older workers are less productive than their younger peers, and ageing didn’t suddenly become an issue in 2009. At the same time, servicing the growing populace of elderly ties up ever more capital that could otherwise be used to boost productivity. The NHS spends about 7 times more money on a man in his mid-80s than it does on a man in his 30s.

Excessive indebtedness in all economic sectors (the government, the household and the corporate sector) is another reason. Think of it this way. The more debt we are saddled with, the more capital that could (and should) be used to enhance productivity is, instead, used to service existing debt.

In other words, capital is increasingly being misallocated; i.e. it is being used for purposes that do not affect GDP growth meaningfully. I am sure my ageing father would argue that his medicine paid for by the government is definitely not a misallocation of capital but, in economic terms, it is.

One simple way to measure capital misallocation is to look at how much GDP grows for every dollar of debt we add. Back in the 1950s and 1960s, before the world fell in love with debt, it was quite normal for the two to grow 1:1, but not anymore. A highly indebted country like the US can only produce about $0.25 of GDP growth for every dollar of new debt. An even more indebted country like China is running at about $0.20 of GDP growth now for every dollar of added debt (Source: MacroStrategy Partnership LLP)

Commercialisation of fusion energy

The other possible, and most likely the ultimate, solution to the productivity conundrum is commercialisation of fusion energy. Now, what is simple in principle, is not always that simple in practice. Scientists thought for many years that traditional nuclear power provided the answer to our energy problems, but various safety issues have never been satisfactorily addressed. Consequently, the technology has never been fully endorsed by the public.

That said, there is another nuclear energy form in the offing. Instead of separating nuclear particles, as you do in nuclear power plants today (a technology called fission), you make nuclear particles collide, and that technology is called fusion.

Fusion is the most basic form of energy in the universe. It is what powers the sun and the stars, where energy is produced by a nuclear reaction in which two atoms of the same lightweight element, usually an isotope of hydrogen, combine into a single molecule of helium.

When scientists attempt to replicate that process, the most important ingredients are sea water and lithium, both of which are in ample supply; hence vast amounts of energy could be produced at a very reasonable cost - at least theoretically. Even better, the fusion process does not suffer from all the safety issues that accompany traditional nuclear power. So far so good, but there is a problem – and a big one at that.

Researchers can produce plenty of energy from fusion but not in a controlled way. The best example is the hydrogen bomb, where a huge amount of energy is released in a highly destructive manner. If the same amount of energy could be released gradually – in a controlled manner – we would have found the eternal solution to planet Earth’s energy requirements.

We would have virtually unlimited access to cheap energy, and greenhouse gasses would be a thing of the past. There would be little nuclear waste, and productivity would rise dramatically across the world, effectively dealing with the debt overhang. These factors in combination would resolve some of the biggest challenges mankind is faced with today. The greenhouse gas problem could also be addressed by accelerating the implementation of renewable energy forms, but that wouldn’t address our desperate need for cheaper energy, as renewables are still comparatively expensive.

Having said that, creating a controlled fusion reaction has proven very difficult. Because the nuclei have the same charge, they will electrically repel each other. To overcome the natural repulsion of the nuclei, you must give them sufficient energy. That means heating them up to about 12 million degrees but, as you heat a gas or plasma up, it expands and the atoms move further apart.

The trick is to contain the heated plasma long enough that the nuclei have the chance to collide and overcome the repulsive force. Researchers have now reached that point and have achieved energy breakeven, but there is still a long way to go, before the technology can be rolled out commercially.

In part III, I will revert with an up-to-date briefing on how the research programme on fusion is progressing.

Why it is so important to boost productivity

If productivity continues its downward trend, it is only a question of time before it turns negative. With workforce growth already being non-existing (and even negative in some countries), if productivity growth turns negative, GDP growth will most definitely turn negative, and so will longevity. Why? Let me explain.

The Industrial Revolution enabled the global population to expand dramatically. At the height of it about 200 years ago, there were only 1 billion people in the world, and now there are 7.5 billion.

The productivity growth of the Industrial Revolution created capital of which human capital was part. If the economy begins to shrink, then the capital stock will also shrink, with human capital part of that.

Longevity has already begun to decline in parts of the US and the UK, as we can no longer afford the best - but also the most expensive - treatment forms. Access to very cheap energy could turn into an important lifeline in more than one sense.

If governments around the world were half as smart as they claim to be, research budgets into commercialising fusion would be multiplied. It is by far the best medication for a global economy in decay. Fusion will almost certainly have a much more dramatic effect on productivity and hence on GDP growth than anything automation can ever deliver.

However, it is a race against time. The global economy could quite possibly sink well before fusion reaches a state where commercialisation becomes viable, a point we won’t reach for many years, but much more about that in part III.

Niels C. Jensen

8 October 2019