Monthly Macro & Markets: March 2023
After a big jump in many economic data points for January, economic indicators broadly weakened in February/March, highlighting how much of January's big jump was just seasonal noise.
Welcome!
For those who are viewing the Monthly Macro & Markets (MMM) report for the first time, the purpose of this report is to provide a comprehensive summary of the economic & market data over the past month, including my personal analysis & breakdown of what it all means, and what lies ahead.
The MMM report is thus your one-stop shop to understanding all things Macro & Markets, and forms a key pillar of achieving my mission, which is to help democratise institutional grade research for the benefit of all individuals, not just large asset managers and high net worth individuals.
If you would like to support my work, please subscribe to Economics Uncovered if you haven’t already, and like and share this latest report — thank you!
This month in macro
A recap of SVB’s collapse & why bank deposits have suddenly become a widely watched metric
Following the collapse of Silicon Valley Bank (SVB), those interested in the world of economics and finance have significantly increased their focus on one metric: bank deposits.
While total US commercial bank deposits have been falling since November 2022 on account of the Fed’s QT and aggressive interest rate hikes, it wasn’t until SVB’s collapse that investors increased their focus on this important metric.
Since turning YoY negative in November 2022, the rate of decline in US commercial bank deposits has accelerated. As of 22 March, YoY commercial bank deposits are down 4.2%. This is far and away the largest decline in the history of the Fed’s current data series, which dates back to 1974.
There’s a good reason that declines in deposits are a rare phenomenon — a leveraged, fractional reserve banking system, encourages artificial economic growth via the expansion of credit.
With banks only holding a fraction of their loan book in cash deposits, an increase in deposits can be multiplied many times over in new credit, expanding the money supply and nominal economic growth.
While such a system generally functions fairly well when deposits are expanding, its leveraged nature means that it struggles greatly when deposits fall.
When deposits fall, banks need to sell assets in order to meet customer withdrawals.
With its tech customers bleeding cash and drawing down on their deposits, this is what occurred at SVB.
With only ~6% of its assets held in cash and cash equivalents, in order to help meet its customer withdrawals, SVB sold some of its fixed income securities.
Though with interest rates rising, and without interest rate hedging in place, SVB incurred a ~$1.8bn after-tax loss on the sale of these securities.
The loss recorded on these bonds meant that SVB then sought to raise additional funds from shareholders to increase its equity position. This led to a huge share price decline, which snowballed into a drastic loss of confidence, and a run on deposits that led to SVB’s swift demise.
On 9 March a total of $42bn of deposits were withdrawn from SVB, and it ended the day with a negative cash position of $958m.
For SVB, that was the end of the road.
The Fed’s actions treat the symptom, not the disease
In order to avoid banks incurring losses on the sale of bonds in order to meet liquidity needs, the Fed created the Bank Term Funding Program (BTFP), which allows banks to lend eligible collateral to the Fed at par.
While this fixes one problem, it only treats one symptom, as opposed to the disease itself.
For the underlying problem isn’t the decline in bond prices, it’s that current monetary policy conditions continue to encourage deleveraging and falling bank deposits. Should this continue, other issues are likely to arise over time.
An additional problem that could arise from the significant decline in bank deposits, is that banks reduce their overall lending. This is likely to further weaken nominal economic growth, leading to banks further tightening their lending standards, leading to a further reduction in credit growth.
This would be a major development, as lending has so far continued to grow at a brisk pace — though it has recently shown signs of turning lower.
A tapering in lending growth, and the potential for outright declines in lending, would further exacerbate the current YoY declines that are being seen in the M2 money supply, which on an annual average basis, is currently running at the most negative pace seen since the Great Depression.
Further significant declines in the M2 money supply would greatly increase the odds of a future deflationary bust occurring.
Final Q4 2022 GDP number is revised slightly lower
After having written extensively on Q4 GDP, and how a deeper analysis reveals a weaker underlying picture than the headline number suggests (see here), the 3rd and final estimate of Q4 GDP saw another revision lower.
After an original estimate of 2.9%, headline real GDP was revised from 2.7% to 2.6%.
A better measure of underlying economic growth, being real final sales to domestic purchasers, which excludes net exports and volatile private inventories, was also revised slightly lower, but once rounded to one decimal place, remained at 0.7%.
CPI continues to trend lower — particularly ex-lagging shelter
With the M2 money supply seeing some of its largest declines since the Great Depression, and real final sales to domestic purchasers falling to low levels of growth, it is not surprising to see CPI growth continue to moderate.
Indeed, February marked the 8th consecutive month of lower YoY CPI inflation, which has now fallen to 6.0% (from 6.4%). Core CPI fell for the fifth consecutive month, reaching 5.5% (from 5.6%).
When the impact of lagging shelter costs are removed from the equation, the disinflationary picture becomes even clearer.
CPI ex-shelter has fallen from a peak of 10.8% in June 2022, to 5.0% in February. Core CPI ex-shelter has fallen from a peak of 7.6% in February 2022, to 3.7% in February.
But the Fed erroneously focuses on lagging “supercore” prices — which remain high
Instead of seeing the clear disinflation that has been occurring for many months, and understanding the nature of the current price cycle, the Fed instead focuses its attention on the most lagging price indicator — core services prices.
As of late, the Fed has rightly also begun to separate out the lagging shelter element of services prices, hence the new term of “supercore” services prices.
On this basis, one can see why the Fed is still concerned about inflation, with MoM price growth once again shifting higher to be above its historical average in January and February, with the gap also growing larger in February’s data.
This also includes the benefit from indirectly measured health insurance prices, which after its most recent annual adjustment, is contributing to significant downward pressure on the CPI in a manner that is not present in the PCE Price Index.
If this element is excluded, one can see that the MoM growth in “supercore” services prices remains well above its historical average, and has been so in all months bar one, since January 2022.
The reason that the Fed’s focus on supercore services prices is erroneous, is that it ignores the manner in which prices evolve.
Namely, the first thing to move is the money supply. Focusing on the current price cycle, the next thing to move were durable goods prices, as individuals spent stimulus checks on items like electronics and home appliances as they were cooped up at home on account of COVID restrictions.
Next, nondurables prices rose as movement and economic activity normalised. Finally, services prices, which are ordinarily the most lagging, began to rise, after having their lag further reinforced by the delay it took for the newly printed money to reach the services sector on account of COVID impacts.
Understanding this price cycle, shows that focusing on services prices is the worst metric of all, for it ensures that the Fed will be too slow to react to rising inflation (as it was), and it ensures that it will be too slow to react to falling inflation (as it now is).
311k nonfarm payrolls added in February, but the signs of a slowing employment market are growing
After 504k nonfarm payroll jobs were estimated to have been added in January, another 311k were estimated to have been added in February, with 3-month average growth remaining well above its pre-COVID average.
While this suggests a strong jobs market, there are several factors that point to a slowing employment market.
These factors include:
moderating private sector job growth — while still from higher levels, there’s an ongoing downtrend in private payroll growth, with recent monthly jobs growth significantly supported by government jobs, which outside of the peak COVID rehiring period, are around peak levels.
a deceleration in the BLS' private sector employment diffusion index, which dropped 12 points in February from 68 to 56 — this indicates that private payroll growth is coming from a smaller subset of private industries.
falling manufacturing employment in February, with the 3-month average growth rate falling to just 5,000.
the BLS' manufacturing employment diffusion index turning negative in February.
Fed branch surveys showing an ongoing weakening of key manufacturing & services sector indicators — in addition to moderating employment indicators, more leading indicators like new orders & hours worked point to further moderation ahead. I elaborate further on these surveys below.
an ongoing moderation in average hourly earnings growth.
an ongoing moderation in the rate of quits — while from elevated levels, and there was a small rise in February, this indicates that individuals are growing less confident of finding better opportunities elsewhere, or are receiving less interest from competitors to move firms.
an ongoing moderation in job openings — while from elevated levels like quits, the most recent data for February showed a material moderation in the level of job openings.
Retail sales decline in February, Redbook data suggests further weakness ahead
Last month, I spoke of how many economic data points were likely to have been unusually strong on account of seasonal noise, and how as a result, it was particularly important to focus on the broader trend.
Much of February’s data confirmed that this was the right approach to take.
For instance, after seasonally adjusted retail trade sales jumped 2.9% MoM in January, and many thought that alongside a bumper January jobs report and higher services prices, that the Fed needed to significantly increase its tightening, retail trade sales fell 0.1% MoM in February.
Just as it did in January, the 3-month moving average of non-seasonally adjusted retail trade sales continues to moderate on a YoY basis, falling below 5% to 4.7%.
Furthermore, in what is a significant negative readthrough for March’s retail sales data, the more timely Johnson Redbook Index of weekly retail sales, has shown a significant further moderation in March.
While there was a moderate increase in the last week of March, the YoY growth rate of the Redbook Index took a material leg lower in March, falling from 5.3% on 28 February, to 3.7% on 1 April.
During March, the YoY growth rate fell to as low as 2.6% in the week to 11 March.
Beyond January’s seasonal noise, real PCEs return to a weak reading
As was the case with retail sales, real personal consumption expenditures (PCEs) saw a significant increase in January. Though as I pointed out last month, there were clear indications that seasonal adjustments were distorting the data.
With January’s large seasonal distortion now in the rearview mirror, real PCEs fell again in February — the third MoM decline recorded in the past four months.
Given the large jump in January, and the relatively large fall in November that rolled out of the 3-month average, the YoY growth rate rose in February and could do so again in March, when December’s decline is rolled out of the 3-month average.
Fed branch manufacturing surveys weaken further in March
Continuing its ongoing deceleration, Fed branch surveys of manufacturing activity saw another significant decline in the new orders and hours worked components in March, which fell to -17.6 and -12.0 respectively.
The employment component eked out a small gain, but remains around 0, rising from 0.3 to 0.6. Despite the overall average gain, 3 of the 5 Fed branches recorded a negative value for the current employment component, which aligns with the BLS recording a decline in manufacturing employment in February.
US manufacturing industrial production continues to weaken
The ongoing weakening being seen in the Fed branch surveys of the manufacturing sector, coincides with a further weakening of US manufacturing production in February, as measured by the Fed’s Industrial Production Index.
The 3-month moving average YoY growth rate has now fallen to a breakeven position, a material weakening from the growth rate of 3.1% that was recorded in October 2022.
Fed branch surveys of the services sector also continue to weaken
In addition to the weakening being seen in the manufacturing sector, the Fed’s surveys of services sector activity also continue to weaken.
While not as far progressed as the weakening that has been seen in the more cyclical manufacturing sector, again the key components are either negative or close to it.
The average of the Fed branch surveys saw the sales index again turned negative in March, the hours worked component fell to 0.3, the part-time employment index turned negative (falling to -3.8), whilst the full-time employment component plunged 6.4 points, falling to 1.7.
Existing home sales rise in March following February’s jump in pending home sales
With pending home sales seeing a significant MoM increase of 8.1% in February, existing home sales jumped 14.5% in March.
While pending home sales were boosted in February on account of four consecutive months of declines in 30-year fixed mortgage rates, a 4.6% bounce in monthly average mortgage rates across March, saw MoM pending home sales growth moderate significantly. This suggests that the large jump in March’s existing home sales may taper significantly in April.
Even with the large MoM increase in existing home sales, it’s important to remember that this bounce comes after 12 consecutive MoM declines, meaning that existing home sales rose from a very depressed level.
Federal government tax receipts continue to indicate a weakening economy
As noted in last month’s MMM report, federal tax receipts have been seeing large YoY declines. This trend continued again in February, with YoY tax receipts down a significant 9.6%. On a 3-month moving average basis, the YoY rate of change remains materially negative, at -6.2%.
Monthly tax receipts have now been materially YoY negative for four consecutive months, providing another indicator of a weakening US economy.
As January’s seasonal noise evaporates and weaker data from February & March rolls through, the Atlanta Fed’s GDPNow forecast is moderating significantly
While many were swept by the noise of heavily seasonally adjusted data in January, as more and more data points for February and March have begun to flow through, the Atlanta Fed’s GDPNow estimate has been revised lower, falling from a peak of 3.5% on 23 March, to 1.5% on 5 April.
With Johnson Redbook Retail Sales data providing a weak readthrough for retail sales and by extension PCEs in March, there’s reason to suggest that the currently relatively significant estimated increase in PCEs of 3.4% (and which the GDPNow estimate had earlier forecast at 5%), will continue to be revised lower as additional data rolls through.
This month in markets
Stocks power ahead — the Nasdaq records a quarterly gain of 16.8%
After markets declined in February, all three US indices recorded gains in March. The Nasdaq recorded a particularly strong gain, rising by over 6%.
All three indices thus brushed off the collapse of SVB, with the S&P 500 and the Nasdaq recording particularly strong gains during 1Q23 — the latter saw particularly strong growth, rising 16.8%.
Markets supported by earnings estimates that continue to not price in a recession
One factor helping to support equity markets, are analyst estimates that continue to not only not price in a recession during 2H23, but which are instead forecasting a huge earnings boost — analysts are forecasting a 7.3% and 13.4% YoY increase in earnings in 3Q23 and 4Q23 respectively.
This comes despite the many indicators that point to a weakening economy, and the delayed impact of the Fed’s aggressive tightening.
On a calendar year basis, analysts are currently forecasting YoY earnings GROWTH — a far cry from what would typically be seen during a recession.
While equity analysts continue to ignore the warning signs, rates traders are now forecasting several 2023 rate cuts
While equity analysts continue to forecast earnings growth in 2023, rates traders aggressively shifted their expectations for future Fed policy following the collapse of SVB.
At the end of February, the market ascribed a 73.2% probability of the Fed setting the fed funds rate at 500-525bps at its May meeting. At the end of March, the market ascribed a 48.4% chance of that outcome occurring, instead slightly favouring the Fed holding rates at 475-500bps.
Further down the road, one can see than an even more significant shift has occurred. For the December meeting, the most likely fed funds rate has been reduced from 500-525bps, to 400-425bps — i.e. at the end of March, the market moved to price in several interest rate cuts during 2023.
Given that the Fed continues to state its expectation for no rate cuts in 2023, a market forecast for several rate cuts would likely imply an economic recession — which would appear to be at significant odds with analyst expectations for booming 2H23 S&P 500 earnings.
Bond prices rise to reflect the drastic revision to rate hike expectations
Given the drastic movement in rate expectations, bond prices rallied during March, which means that their yields fell.
The biggest movement was seen in the 2-year Treasury, which saw a major drop alongside SVB’s collapse. After closing at a yield of 5.1% on 8 March, the 2-year Treasury ended the month at 4.0% — a more than 100bp move lower in a matter of weeks.
Putting it all together
To summarise the key economic data over the past month, we’ve seen:
a further decline in commercial bank deposits and the M2 money supply;
a further downward revision to Q4 GDP (from 2.7% to 2.6%);
a further deceleration in the CPI (from 6.4% to 6.0%) and the core CPI (from 5.6% to 5.5%);
a further deceleration in the CPI ex-lagging shelter (from 5.7% to 5.0%) and the core CPI ex-lagging shelter (from 4.0% to 3.7%);
that adjusted core and supercore CPI services prices — which the Fed focuses upon — remain elevated;
311k nonfarm payrolls added in February, but many signs of a slowing employment market;
Fed branch surveys of manufacturing and services activity continue to weaken;
retail sales record a MoM decline in February, after a large jump in January. The YoY change on a 3-month moving average basis continues to moderate;
the Johnson Redbook Index of retail sales data indicate a material weakening of retail sales in March;
PCEs fall in February, after seeing a significant rise on account of seasonal distortions in January;
existing home sales jump 14.5% MoM in March, after pending home sales rose by 8.1% in February — though pending home sales growth moderated in March (up 0.8% MoM) and the rise in existing home sales only comes after 12 consecutive MoM declines, meaning activity remains closer to GFC lows than recent highs;
Federal Government tax receipts continue to decline — down 9.2% YoY in February, and down 6.2% YoY on a 3-month average basis; and
a gradual weakening in the Atlanta Fed’s Q1 2023 GDPNow forecast — after peaking at 3.5% on 23 March, it has now fallen to 1.5% as of 5 April.
Putting it all together, shows that economic data continues to suggest a weakening economy. Given the extent of the tightening that has taken place, and with M2 now declining, this is to be expected.
As opposed to January’s seasonal noise, the extent of the tightening that has taken place, and its delayed impact, is the bigger picture that people needed, and still need, to focus on.
While many were getting caught up in the noise of January’s data, sending bond yields and rate expectations higher, by focusing on the bigger picture, I was able to warn that this could represent “the big blow off period”:
“With markets delivering a MAJOR repricing of future hikes off the back of a single month of economic data, is this the big blow off period for rate hike expectations, which could now turn lower over the months ahead? I believe that very well could be the case….
With the strength seen in many other economic data points likely overstated in January due to seasonal adjustments, other indicators like Fed manufacturing surveys continue to point to a downturn.
With the M2 money supply getting increasingly negative, I continue to believe that not only lower inflation lies ahead, but a recession.”
The collapse of SVB and other major US banks has finally resulted in greater attention being paid to falling bank deposits, and the repercussions for future economic activity.
As a result, markets moved to aggressively lower their outlook for the fed funds rate.
While I believe that the bond market is right to price in interest rate cuts from the Fed during 2023, the Fed continues to remain adamant that won’t happen.
Just as the Fed was too slow to react to inflation, it is repeating its previous mistakes in reverse. As opposed to high inflation, the greater risk now is that of a deflationary bust.
Just as prolonged high inflation forced the Fed’s hand and saw it aggressively raise rates, I anticipate that at some point during 2023, a weakening economy will yet again force the Fed’s hand, sending the fed funds rate lower.
In saying that, it’s also important to note that should services prices and/or core inflation remain relatively firm in March, there’s the potential for rate expectations to be revised somewhat higher, as markets potentially become more focused on, and concerned about, the outlook for inflation.
What I’ll be watching most over the month ahead
Over the month ahead I’ll be closely watching the upcoming jobs report for any further indications of a slowdown in employment demand, as well as the CPI report, for any indication that services price growth could be beginning to roll over.
Should services prices and/or core inflation remain relatively firm, there’s the potential for some of the drastic downward movement in rate expectations to move back the other way, with markets once again potentially growing somewhat more concerned about the outlook for inflation.
I intend to soon release my US CPI preview for March, and expect to produce a summary of the upcoming jobs report either here, or on Twitter — stay tuned!
Something that caught my attention
Gold.
With gold holding up relatively well amidst aggressive Fed tightening, with rate expectations now being revised lower, the yellow metal has been on a tear, rising back above $2,000 per ounce. Gold is now closing in on its all-time high that was reached in August 2020.
Given that markets have drastically shifted their focus from the ultimate extent of the Fed’s aggressive tightening, to now being focused on recessionary concerns and when the first rate cut will be delivered, the outlook for gold has become significantly more positive — not to mention the recent surge in global central bank gold buying activity.
Though one potential stumbling block for gold prices, is the potential for another relatively hot CPI report in March. Should services prices and/or the core CPI continue to remain relatively hot, markets could quickly shift their focus back to the outlook for inflation, and whether further Fed rate hikes are needed.
Have your say!
Feel free to share your view on the outlook for interest rates (or inflation, or employment, or the broader economic outlook, or whatever data point has got you thinking), in the comment section below!
To recap last month’s poll, the majority of readers disagreed with the viewpoint that I have been expressing for many months now — being that high inflation will dissipate as M2 declines. As such, I was firmly in the “No: expect major disinflation camp” — a position that I continue to hold.
In case you missed it …
During March, I published five research reports. In case you missed one, or would like to review any report again, here’s a summary of what I published:
Thank you for reading!
I hope you enjoyed reading the latest installment of MMM — feel free to share your opinion on the current macro & market situation below!
In order to help support my independent economics research, which aims to democratise access to institutional grade insights & analysis (as opposed to it being the exclusive domain of hedge funds & high net worth individuals), please like and share this latest research piece.
If you haven’t already subscribed to Economics Uncovered, subscribe below so that you don’t miss an update.
Thanks, that's a very comprehensive review.
With jobs openings, I've heard now that since Covid the numbers are skewed to the upside. If you search for some remote position on LinkedIn, for instance, you might find it advertised as a job openign in a myriad of cities, when really it is just one position.
💯💯