Inflation fears about the US

There are widespread concerns that the large-scale quantitative easing by the US Federal Reserve will lead to a resurgence of inflation. Robert Keavney weighs up the evidence.

I am not a trained economist – one of the many things I am grateful for in this life. I am a three-decade practitioner in financial markets who has noted the empirical failure of many orthodox economic and investment theories. As a result, whenever I confront a theory that I don’t understand – that is, most of them – I have acquired the habit of looking at the data.

Currently, there is a widespread view that the massive ‘printing’ of money will lead to an outbreak of inflation. The theory goes that increasing the volume of money more rapidly than the increase in goods and services, must force prices upward. More money is chasing the same amount of goods, as it is sometimes put.

This has prima facie credibility. If a lot of new money is thrown into an economy it must have consequences, and surely one of them will be upward pressure on prices. But before we accept this idea we must look at the evidence.

To do this we need to understand the concept of the quantity of money in a country. How much money exists in America depends on how you measure it. Certainly notes and coins are money, but are travellers’ cheques money?

And should the value on money market funds be counted as money? Different measures of money (eg, the monetary base or ‘money base’, M1, M2, etc) include or exclude various items.

If many items are included in a measure it is called a broad measure. If few items are included it is a narrow measure. Money base is a narrow measure.

Figure 1 shows that from The Declaration of Independence in 1716 through to 2008 the US money base grew to almost US$1 trillion.

In the three years since then it has approximately tripled – there has been unprecedented money creation.

The Federal Reserve (the Fed) hopes that this will support the weak American economy.

However, it happens that there is little correlation between changes in money base and inflation, as was highlighted in the 1970s.

Annual inflation was more than three times its 50-year average for this decade, while the money base grew at a slower than average rate.

Perhaps we should look at a broader measure of money, say M2, to see if there is a meaningful historical relationship between money creation and inflation. Bank reserves represent a large portion of money base, but they are not counted in M2. It is important, in what follows, not to confuse money base and M2. 

Figure 2 shows rolling annual changes in the consumer price index (CPI) and M2. It transpires that the correlation between the two is only 16.6 per cent.

However, this weak relationship is not unexpected, as common sense would suggest that it would take time for changes in money supply to affect prices. 

If inflation is lagged by three years, allowing three years for changes in money supply to flow into prices, there is a correlation between changes in M2 and subsequent inflation of 54.3 per cent.

This is shown in figure 3, which is a scattergram. Each dot represents CPI and M2 for a particular year. CPI is the vertical axis and M2 is the horizontal axis.

In layman’s terms, a correlation of 50 per cent is halfway between a perfect one-to-one relationship and no relationship at all.

M2 is clearly related to inflation to some extent but not to the extent that changes in one will always be accompanied by changes in the other.

Incidentally if we lag inflation for a greater or lesser period than three years, the relationship is not so strong. 

Where is the cash?

Now we must note some unexpected facts. Over the last half century, rolling annual growth in M2 has averaged 6.9 per cent. But over the three years since the money base took off, M2’s growth has been below average.

How can this be? Where is all the new money? Why hasn’t it hit M2? How can a broad measure of money growth be slower than average after massive money creation using another measure?

Here we must touch on the question of velocity and liquidity traps. The Economist defines velocity as:

“The speed with which money whizzes around the economy, or, put another way, the number of times it changes hands. Technically, it is measured as gross national product divided by the money supply (pick your own method of measuring money supply).”

Dr J Hussman, of Hussman Funds, describes velocity as the dollar value of gross domestic product that the economy produces per dollar of monetary base, or the number of times that one dollar ‘turns over’ each year to purchase goods and services in the economy.

Hussman points out that the Fed’s belief that the expansion in the money base will grow the weak economy relies on an assumption that velocity won’t decline in proportion to the increase in money. He concludes that “this assumption fails spectacularly in the data – especially at a zero interest rate”. 

He supports this with figures 4 and 5, which show money base and velocity in America over 47 years and in Japan over two decades, respectively.

Both graphs show unambiguously that, as money base is expanded, velocity reduces.

In simple terms, the new money doesn’t make it to the economy.

This is called a liquidity trap. History suggests the Fed’s massive quantitative easing won’t produce the desired result. Mr Bernanke is trying to push a river faster than it flows. This won’t affect the river, but it can exhaust the one pushing. 

We have already seen that the recent increase in the money base is not making it to the economy because it has not appeared in M2.

In fact, over the last half-century the correlation between M2 and the money base is close to zero.

So where is the new money produced by quantitative easing?

Sitting on bank balance sheets and in reserves (which, as noted above, are not included in M2).

Why?

Because Americans are deleveraging, not borrowing. It doesn’t matter how much new money banks have available to lend if no one wants to borrow. Increasing the amount available won’t fix the problem, because the constraint is not a lack of available funds.

But this leads us to a question: if the money doesn’t make it to the economy, how can it affect inflation?

Let’s summarise where we are so far:

• The money base is exploding but historically this bears no relationship with inflation or M2; 
• Historically, large expansions in money base do not necessarily impact the economy because velocity falls; and 
• M2 has a moderate correlation with subsequent inflation, but M2 is growing more slowly than average.

At this point it could be tempting to dismiss fears about inflation outright. Nonetheless, there are some qualifications to this conclusion.

Since 1959 M2 has, on average, been nine times greater than the money base, so the latter has been far less relevant to the overall economy.

However, in the last few years this ratio has rapidly changed. M2 is now less than four times the money base. It is not clear what, if any, consequences this change could have.

The last half-century also suggests that neither the money base or M2 ever materially contract. If this continues, the newly ‘printed’ money in the money base will remain there. If so, surely at some time in the future it must find a way into the economy. But when is ‘some time’?

Testing does not reveal any identifiable lagging relationship between money base and M2 (ie, there is no known pattern for expansions in money base flowing into M2).

Finally, Zimbabwe’s experience does suggest that there is a level of money creation that drives inflation to absurd levels – though fantastic money expansion was only one of many bad policies pursued in that country. Presumably, America’s money expansion will never reach those levels.

Inflation – why worry?

Large and sustained changes in inflation have many consequences for markets. The transition to high inflation in the 1970s created a crippling headwind for most asset classes. Conversely the transition back to low inflation in the 1990s created benign conditions for most markets. 

Significant and sustained changes to inflation are rare, but investors must be prepared for them when they occur. This is behind our interest in whether the explosion in the money base could herald a resurgence of inflation.

Conclusion

The Fed has been too active over the last dozen years, and the decisions it has made have been consistently wrong. 

Interest rates were at high settings leading into both of the last two recessions, which contributed to the slowdown. Rates were held so low after the dot.com recession that they contributed to the subsequent mania that culminated in the global financial crisis.

The Fed recently vastly increased the money supply, blind to the reality it would simply result in a lowering of velocity and would not materially benefit the economy.

The situation would have been better if each of these actions had been less extreme.

Undoubtedly there will be consequences from the Fed’s expansion of money. There is no free lunch, and if you conjure approximately US$2 trillion out of the air in three years, there will be a price to pay.

However, they won’t be the consequences the Fed seeks. I fear they will be negative.

Nonetheless, in a failure to recognise the lessons from Japan’s experience, there is a real possibility that there will be further quantitative easing. It would seem that little is being learnt from the past. It makes me glad I’m not an economist (with apologies to Dr J Hussman and all the other exceptional economists).

We must acknowledge that recent events in America are unprecented, including the scale of the expansion in the money base. This must qualify any conclusions we draw from history. However, the fact is that history suggests that an expansion of the money base need not be inflationary. 

This does not guarantee that inflation can’t rise, but it will require other factors to bring it about.

Robert Keavney is an industry commentator.