QT after one year – broad money growth has halved

  • After one year of QT, the Fed’s balance sheet has slimmed by just over $300bn (-6.6%)
  • Although the US government deficit has widened, broad money growth has halved
  • The Fed will still raise interest rates in December; but ‘three hikes in 2019’ is no longer on

Last week, we had US monthly broad money data for October.[1]This also means that we can now gauge American monetary trends after 13 months of quantitative tightening – the Federal Reserve’s slimming of its balance sheet, which began in October 2017. According to the Fed’s projections from June 2017, its holdings of Treasury Securities should after this period have fallen by $196bn. In the event, the actual number from end-September 2017 to end-October 2018 was -$175bn. Overall, the Fed’s balance sheet has shrunk from $4.506tn to $4.208tn, or just short of $300bn. Since the introduction of this policy was phased and the run-off of maturing assets has only now reached its maximum of $50bn per month, the balance sheet shrinkage should be faster over the coming years.

This is much as expected. More importantly, though, what has been the impact of quantitative tightening (QT)? This is a key question. As we have pointed out in previous comments, this is an entirely new situation. While large central bank balance sheets are not new, there is no precedent for a central bank actively shrinking its balance sheet in absolute nominal terms. With other central banks likely to embark on QT over the next few years, the economic impact of QT is obviously relevant when deciding how to proceed and how to formulate monetary policy.

The starting point for the impact of QT is two assumptions. First, that the impact of QE was an injection of liquidity in the economy; that QE pushed down bond yields; and that it pushed up asset prices, leading to a reallocation of assets away from fixed income. The second assumption is that QT will have the opposite and (broadly) equal consequences. This is not necessarily true. QE involved central banks buying assets from the banking sector or from the non-bank private sector. But QT, at least as practiced by the Fed, involves presenting maturing assets for redemption to the issuer – usually the Federal Government. The impact on the economy therefore depends very much on what the government then does. In essence, if the government reborrows the same amount that it has just repaid the Fed, and does so from the banking system, the impact on money, and hence on activity, should be neutralised.

So far, the impact on asset prices is varied. The yield on the 10-year US Treasury was 2.34% at the beginning of October 2017. By the first week of November 2018, it was 3.06%; a rise of just short of 80bps, but also a rise of close to 50%. But share prices are actually up, in spite of recent falls. At the end of September 2017, the S&P500 index closed at 2,519; by mid-November 2018, it was 2,736, a rise of 8.5%, although a fall of 6.7% from its peak on 20thSeptember 2018.

The economic impact is also less clear-cut. On the one hand, the US economy is growing surprisingly strongly. But, with the admittedly significant exception of confidence measures, forward-looking indicators are generally weakening. The housing market is making a distinctly soggy impression, evidenced today by the news that the NAHB Housing Market Index had fallen from 68 in October to a 27-month low of 60 in November. Most importantly, broad money growth has halved. From February to October 2017, US broad money grew in excess of 6% per annum, a rate that over the medium-term is broadly consistent with trend output growth. This then slumped to 1% in the year to February 2018; and has since only recovered to between 2.5% and 3%.

It could have been worse. Over the period October 2017 to October 2018, commercial banks increased their holdings of Treasury and Agency securities by $79bn. 43% of this consisted of mortgage-backed securities, with the rest being non-mortgage backed securities, presumably mainly Treasury bonds. In other words, of the roughly $300bn that the Fed balance sheet shrank by, slightly less than one third found its way back to the banking system, off-setting part of the contractionary impact of QT on money supply. But in a world where bond yields are steadily rising, the attraction for banks to load up on more Treasuries is likely to be limited, meaning that the dampening impact from QT on broad money growth will become more pronounced.

Clearly, then, QT is having a significant dampening impact on US monetary developments, and this will by early 2019 feed through and become visible in lower economic activity. At the same time, however, inflation remains quiescent, giving the Fed the option of putting its current policy on hold. The Fed is still keen to normalise monetary policy and is likely to raise interest rates again in December this year. And it tends to ignore monetary developments. But with other leading indicators also turning – gently – south, the forecast three interest rate increases in 2019 may not materialise after all. This is the more so since the impact of the Trump Administration’s fiscal package will dissipate next year and Democratic control of the House of Representatives will make a repeat of the fiscal stimulus more difficult. Comments by Mr Richard Clarida, the recently appointed Vice Chairman of the FOMC, point in the same direction.

Expect, therefore, US monetary policy to be at least paused from mid-2019 onwards. This will be important for the world economy; because, as we will show in a future Comment, monetary policy elsewhere in the advanced economies, tends to be moving towards tightening.

[1]‘Broad money’ for the US always refers to our own recreation of the M3 measure that the Federal Resereve ceased to publish in 2006. It is a better indicator of economic activity than the narrower M2 or M1 measures.