Inflationary Bank Lending and Money Supply Growth
Given that there is currently no “QE” program underway – with the exception of the reinvestment scheme designed to prevent the Fed’s balance sheet from shrinking (if it were to shrink, the money supply would decline as well) – money supply growth depends primarily on the amount of fiduciary media created ex nihilo by commercial banks.
Putting it differently, it depends on the growth in bank lending, since new uncovered deposit money comes into being by the extension of credit by banks. This deposit money is a money substitute that is only partially covered by standard money, or potential standard money (i.e., bank reserves). However, it has to be regarded as part of the money supply, given that it is used for the final payment of goods and services. From the perspective of its users, it is money.
Photo credit: .Kai
Since the crisis of 2008 and the collapse of the mortgage credit bubble, the following trends have been in evidence: lending to corporations has quickly reached growth rates usually associated with boom conditions. Consumer lending has by contrast been more subdued, with mortgage credit growth not surprisingly only very slowly moving back into positive territory. Most of the acceleration in bank lending could be observed once “QE” was tapered and ended – as a result, broad US money supply growth has remained brisk, even though it is far below its peak levels of recent years.
Since monetary inflation is the most important factor propping up malinvested capital in the economy as well as assorted asset bubbles, it is worth keeping a close eye on bank credit growth at the moment, as it should lead changes in money supply growth rates. The effect monetary inflation exerts on prices in the economy is unevenly distributed, both across goods and services and across time. For instance, a central bank can get away with a lot of money supply inflation as long as people believe the policy will one day be stopped or reversed. When this belief wanes for some reason, prices will tend to rise very quickly. So there is no one-to-one relationship between money supply growth and prices. Leads and lags and the uneven distribution of price changes are major characteristics; they depend on the demand for money (cash balances), resp. on the discrete points at which new money enters the economy.
Credit Growth Begins to Slow Down
Since the biggest price effect of the inflation of the money supply in recent years has been on assets such as stocks, bonds and real estate, these are the sectors that will be most susceptible to a slowdown in the pace of inflation. In light of current valuations, it could well be argued that asset prices are unlikely to rise much further unless inflation actually accelerates. However, annualized growth in bank credit has actually begun to slow recently. The slowdown is still small, and may yet reverse – but given the economy’s sluggishness, what reason is there for banks to extend more credit and increase their risk? Why should potential borrowers increase their credit demand?
It appears to us that a lot of the credit created in recent years has either fed malinvestment (e.g. in the oil patch) or financed financial engineering, with listed companies funding stock buybacks, M&A activity and in some cases even dividend payments on credit (much of this has been obtained via the bond market, but there are in turn many bond market investors using leverage, which is funded by banks). As Zerohedge recently reported, a sharp slowdown in free cash flow generation may be forcing a rethink with respect to this strategy.
As an aside to this, it is quite amusing to us that record lows in buybacks invariably coincide with times when stocks are actually cheap, while record high buybacks always occur when stocks are already horrendously overvalued and often near a major peak. Shareholders are usually happy when corporate chieftains announce big buybacks (which they do mainly for their own benefit), but they really should think twice about this. It often turns out to be a terrible waste of capital.
Below are several charts illustrating the current situation in bank lending growth. The first chart shows commercial and industrial loans and their annual growth rate over the long term (so as to illustrate what growth rates are usually associated with booms and busts).
Commercial and industrial loans. The current annualized growth rate of 10.87% remains consistent with boom conditions – however, it does represent a slowdown from the secondary peak just above 12% annualized attained in 2014. Interestingly, these growth rates tend to peak before recessions begin, but bottom after recessions end. So C&I loans are a leading indicator of busts and a lagging indicator of new booms. Note that the big upward spike in 2008 was the exception from the rule: at the time companies desperately drew down their credit lines, as bond market funding froze and they feared banks would unilaterally cut their credit lines – click to enlarge.
Next we take a look at total loans and leases, i.e., commercial and consumer loans together (excl. mortgage debt). The growth rate of this category has been slower than that of C&I loans, due to the comparatively slower growth in consumer loans. Consumers are still reeling from the after-effects of the burst real estate and mortgage credit bubble and their eagerness to buy things they don’t need with money they don’t have in order to impress people they don’t like is evidently much reduced. Here too we see a slight slowdown in growth to a recent 7.53% – this presumably largely reflects the slowdown in C&I lending.
Total loans and leases and the annual growth rate. Note that previous post-crisis slowdowns in bank lending were more than compensated for by massive QE on the part of the Fed – however, currently any slowdown in inflationary lending should also lead to a slowdown in money supply growth. Until “QE4”, that is – click to enlarge.
Next we take a look at consumer loans. In this case we show the annual growth rate in a separate chart, as there was a change in accounting rules in 2010 which brought several 100 billions in loans that were previously held off-balance sheet back onto bank balance sheets. As a result there were wide short term swings in the annualized growth rate, making a combined chart a bit hard to read.
Also, when considering the total stock of consumer loans, one has to mentally adjust for this one-time addition (which is easily visible on the chart). When adjusting for this one-time jump in the series, one realizes that consumer loans actually remain slightly below their 2007 peak level. They are growing again, but at 4.76% their growth is quite sluggish compared with historical levels. The average US consumer’s balance sheet has been thoroughly eviscerated by the preceding credit bubbles, particularly the housing bubble. One can thank prominent Keynesian economists like Paul McCulley and Paul Krugman for propagating the latter and Alan Greenspan’s Fed for implementing their suggestions. See Mr. Krugman’s comments back in 2002, when he was still against large budget deficits (possibly because the president was a Republican?), before he became a proponent of even bigger deficits. The money quote:
“To fight this recession the Fed needs more than a snapback; it needs soaring household spending to offset moribund business investment. And to do that, as Paul McCulley of Pimco put it, Alan Greenspan needs to create a housing bubble to replace the Nasdaq bubble.”
Mission accomplished. The 2001 recession was successfully “fought” at the small price of the biggest financial and economic crisis since the 1930s striking 5 years later. This should actually be seen as a general warning against attempts by monetary bureaucrats to “fight recessions”. The outcome is always the same: more debt, even bigger bubbles and even bigger busts.
Consumer loans, annual rate of growth. In boom times, this growth rate regularly spiked above the 10% mark. Note that the even bigger spikes in the 70s and 80s are partly connected with then much higher consumer price inflation rates. Recent growth of 4.76% is positively anemic considering we are in the 7th year of a recovery – click to enlarge.
Lastly, here is a chart of total mortgage debt. Its annual growth rate is recovering, but at slightly over 2% remains well below the growth rates observed at previous recession troughs. Without massive government intervention – the FHA and the agencies have essentially brought sub-prime lending back at full blast, with down-payment, income and credit score requirements no private lender would even remotely consider anymore these days – we doubt that positive growth in mortgage lending would have returned so quickly. The astonishingly strong echo bubble in house prices over recent years probably guarantees that another bust will sooner or later strike in this sector (we believe that one of the reasons why the Fed is so reluctant to hike rates is that the ongoing credit bubble is held to be quite vulnerable to a change in conditions).
Total mortgage debt and its annual growth rate. Mortgage credit bubbles have been a major and regular feature of US economic booms, and the absence of one in the current recovery partly explains why even aggregate economic statistics measuring “economic activity” are so weak – click to enlarge.
Money Supply Growth – Will it Slow Down Soon?
There are lead-lag relationships between the growth in bank lending and the growth of the broad money supply measure TMS-2, but they are difficult to clearly discern at times, as they depend on who is doing the inflating, as well as on the flow of dollars to and from dollar deposits held abroad (any dollars held abroad won’t be counted in domestic money supply statistics). In “normal” times, the Fed can also push up the money supply quasi-permanently by means of coupon passes – outright purchases of securities instead of repos that are reversed at the end of their term.
When the Fed purchases securities outright, such as it has done in its QE operations, both bank reserves on the Fed’s balance sheet and deposit money in the economy increase. Legally, the primary dealers are considered non-banks, so once they deposit a check they receive from the Fed and the bank redeposits it with the Fed, both deposit money and bank reserves increase (for an in-depth discussion of this mechanism see “Quantitative Easing Explained”).
Anyway, given the current situation – there is no special incentive to shift US dollars from the euro dollar market to domestic US accounts (such as was e.g. the case during the euro area debt crisis) and the Fed isn’t engaged in QE – it is fair to assume that changes in bank lending growth will lead changes in money supply growth. The first chart shows the annual growth rate of TMS-2 compared to growth in all bank loans and leases. In this chart there seems to be no rhyme or reason with respect to the above mentioned leads and lags, as it doesn’t include mortgage debt. We mainly show it because we suspect that the major component of loans & leases – commercial and industrial loans – is the main driving factor in credit growth trends at present.
At the end of August, TMS-2 growth stood at 8.55% while growth in total loans and leases had slowed to 7.54%. TMS-2 growth has essentially moves sideways since the tapering and end of QE3, as bank lending growth has made up for the lack of direct monetary pumping by the Fed – click to enlarge.
Next we look at the same date, but with mortgage debt added – i.e., this chart shows the combined rate of growth of all loans and leases plus mortgage debt. Given the large size of total outstanding mortgage debt, its growth rate influences changes in the overall credit growth rate quite strongly. These data are not as fine-grained and up-to-date as those in the above chart, as mortgage debt is only published on a quarterly basis. So the situation depicted is as of Q2 2015.
Here we can see that total credit growth had not yet slowed down as of Q2 – mainly because growth in mortgage lending has resumed, even if it remains at a historically very subdued level – the higher base is the decisive factor in this case.
Next we take a look at a narrower gauge of the money supply, namely M1 (currency and demand deposits), which is close to the narrow Austrian money supply TMS-1 (sweeps would have to be added to it, but the Fed no longer publishes sweeps – moreover, their importance has declined in light of the vast excess reserves held by banks. Sweeps were mainly a means of circumventing/ emasculating minimum reserve requirements, which allowed for an acceleration of bank credit inflation during the boom of the 1990s).
M1 growth rates are more volatile than those of TMS-2, due to the comparative inertia of the large savings accounts component of the latter, but a trend is nevertheless clearly discernible. Conceptually it makes sense to look at this narrower gauge of the money supply, because this is the portion of the money supply used for the bulk of final payment transactions in the economy. While savings accounts are de facto available on demand (de iure the situation is less clear, as banks can theoretically impose a withdrawal delay of up to 30 days), they are as a rule not employed in day-to-day transactions apart from less frequent big-ticket purchases.
As can be seen, annual growth of M1 has slowed down considerably since its QE-induced peaks. The pace of the slowdown has been mitigated by the resumption in bank credit growth as well, but the overarching downtrend clearly remains in place.
As long as there is no new QE program, future trends in bank lending growth will be decisive in determining money supply growth. We will continue to post updates on this, especially if any large changes occur. The entire bubble edifice is highly dependent on money supply growth, so we believe this information is quite important for traders and investors.
Charts by: Saint Louis Federal Reserve Research