Interest rate versus economic growth

The following is an excerpt on Monetary Policy from a paper I wrote in May 2023:

It is now more than a year since central banks began raising interest rates. The Fed started in February 2022, and the ECB followed in July 2022. However, despite interest rate hikes of 500 basis points and 325 basis points, neither the US nor the EU are in recession. In particular, neither is there a deep recession, as predicted by the macroeconomic models. The market is divided on why this is the case. The answer could significantly impact the future monetary policies by the Fed and the ECB.

Mange klassiske pengepolitiske “regler” er sat ud af kraft

Initially, it is worth remembering that since the financial crisis, many other classic and neoclassical economic relationships have also deviated from theoretical predictions. Previous writings (November 2022) outlined central banks' concerns about the shifting correlation of the the Phillips curve Since the financial crisis, inflation appeared to be unaffected by employment. Then Covid lockdowns and stimuli suddenly caused inflation to surge. However, today inflation is falling again, even though employment is still rising, and there is a labor shortage in many places in the USA. In other words, the Phillips curve's prediction is once again challenged.

The paradox is compounded by the fact that the economy is still growing despite unusually large risks from geoeconomics, geopolitics, and supply chains. These should have increased inflation because companies are consolidating (e.g., business inventories in the USA). Furthermore, the stock market expects economic growth to rise in the coming years, with indices near all-time highs. The debt market also expects the Fed to start cutting rates this year. The Fed itself contrarily expects at least to maintain the rates.

Central banks' interpretation of the causes of the interest rate paradox matters. This decides their future monetary policy, but it also affects politicians' perception of where the economy's limits and logic lie. Up until 1½ years ago, large parts of the political spectrum in the USA, for instance, argued for MMT. If high interest rates do not curb economic growth, only the liquidity instrument remains.

Rules and relationships are based on ceteris paribus assumptions

The greater the debt in the system and the closer the economy is to full employment, the greater the risk of imprecise monetary policy dosing. An overheated economy risks collapsing instead of cooling down.

Over the past year, there has been a significant difference between what the major banks' macro models predicted that interest rate hikes would do to the economy and what the Fed's own models suggested. The Fed's predictions have been clearly more accurate, and therefore the bank has retained a steady hand despite a delayed start. However, it is fragile to be the (almost) only one who has been reasonably accurate on predictions. Usually, one is never better than the accuracy of the latest prediction. Trust takes a long time to build but a short time to damage. Therefore, it is fragile when only one of the major players continues to get it right.

In understanding the interest rate paradox, it is worth considering the differences between the models' structures and assumptions. There has been disagreement on:

• The time lag between an interest rate hike and when the economy is noticeably affected
• The real effect of interest rate hikes on the economy
• The significance of interest rate versus liquidity instruments, including QE vs. QT

In part there is a time lag

The time lag,"Transformation lag," refers to how long it takes from the interest rate hike until a measurable effect on the economy occurs. This is difficult to determine for several reasons.

• The lag can be measured, for instance, as when the peak of the effect occurs and as when the total effect occurs.
  • The peak was calculated by the BoE in 1999 to around 12 months.
  • The total effect was calculated by the Czech National Bank in 2013 to be 29 months. For developed countries this was up to 50 months. The higher the growth and debt, the shorter the time before there is generally a full effect.
• Estimates like the above are used as norm figures for mathematical models.
  • In the real world, economies are also affected by many other external factors. For example, the more open an economy is, the more it depends on the financial and economic developments of other countries.
• Therefore, can the effect be measured at all? Over a period of 12-50 months, many geopolitical and innovation-related changes also occur. They also impact a country's economy.
• In the current interest rate paradox, the time lag is an incomplete explanation. This is because it has been over 15 months since the Fed started raising rates. The peak should therefore have occurred already, or a ketchup effect should be underway. Instead, we see the opposite. For instance, employment is still rising.

In part the effect is difficult to isolate

The effect is also difficult to determine because the economy is influenced by many other factors. In addition, models suggest that the effect comes in waves, i.e., a rapidly rising effect that "fades" into a "long tail." The effect also depends on the starting point and where and how it is measured.

• For example, if the economy has low valuations of real assets and financial assets, an interest rate hike should mean less for economic growth. Similarly, a decline in real asset values should be less critical. This is because the part of private consumption based on home equity would be less affected.
  • But today, real assets are very high relative to wages. There is a "housing affordability crisis" (housing bubble).
  • Therefore, the effect of interest rate hikes should have been significant. It erodes a relatively larger part of the potential for private consumption.
  • But we see the opposite right now. Private consumption is stable, and in the USA even slightly increasing. This is also the case for discretionary spending such as restaurants and takeaways. Sensitivity has been significantly lower than expected.

And in addition there is a timing problem

• Furthermore, timing and effect are also influenced by institutional factors such as "windows" for refinancing. For it is only then, that a real market effect of interest rate hikes can be seen.
  • In the UK, for example, the "window" for when mortgages can be refinanced creates a lot of bias.
  • Therefore, there is concern about whether a ketchup effect could occur in the UK. The perfect storm of an economy in decline from Brexit, etc., combined with a housing bubble hit by a sharp rate hike but only triggered in connection with the refinancing "window."
  • So far, the extent of loans in the refinancing window seems too small to trigger anything major.

Finally, it is difficult to determine which monetary policy contributes the most

Finally, it is difficult to separate how much interest rate versus liquidity instruments matter for the economy. Over the past 10 years, the outlook for QE, i.e. increased liquidity, has impacted markets significantly more than interest rate announcements.

The logic behind this is more straightforward. There is a very short "transformation lag" if liquidity increases. This raises general bid-to-cover ratios for bond issuances, for example. Supply versus demand increases prices and lowers interest rates, and consequently, this also raises stock prices. The higher the debt in the market, the more sensitive the financial stability of the market becomes to whether liquidity is always abundant. One can survive with a deficit, but not with a lack of liquidity and credit.

But macro economically, there is insufficient historical data for the liquidity effect. This is because QE wasn't introduced until in 2009, and in the EU in 2015. Since then, central bank balances in developed countries have quadrupled, and economies and employment have only moved upwards. But interest rates have been calmly and steadily falling. They have even been close to zero for the last 10 years. Until 1½ years ago. Therefore, it is difficult to justify the liquidity effect with sufficient empirical data. We may be able to argue for the logical connection, but we cannot methodically demonstrate the correlation.

Is it because there is still ample liquidity (overall), or ...?

So far, the Fed's and ECB's QT has been moderate and steady. There are furthermore no signs of a liquidity squeeze in the market, either as a whole or in specific sectors. The collapse among regional American banks had individual causes, not market structural ones. Neither are there significant declines in bid-to-cover ratios for bonds. Finally, Reverse REPO and total bank lending remain stable. Drop in the M&A market and IPOs can still be explained by the fact that credit quality has started to cost. This causes multiples on uncertain quality to fall.

Therefore, there is no clear answer on why interest rate hikes have not led to a recession. Neither is there an indication whether a "ketchup" recession is on the way. There is no simple explanation for why the previous macro models are wrong in their predictions.

But it is fundamentally concerning that many of the classic economic relationships are currently showing such ambiguity. For if one has to curb an overheated economy, there is a risk of collapse.