Connect with us

Economics

UMich Sentiment Slides In January As Inflation Expectations Hit 10-Year Highs

Current Conditions are the worst since September 2011…

Published

on

This article was originally published by Zero Hedge

UMich Sentiment Slides In January As Inflation Expectations Hit 10-Year Highs

After a small bounce in December – from decade lows – University of Michigan Sentiment was expected to slip back lower in preliminary January survey data, and it did… more than expected. Both current conditions and expectations fell in the flash January data, dragging the headline print down from 70.6 to 68.8 (below the 70.0 expectations)…

Source: Bloomberg

Current Conditions are the worst since September 2011. Current conditions in the national economy were reported to have worsened by half of all households in each of the past three months, and just one-third of all households have anticipated improved economic conditions during the year ahead.

When asked to evaluate the year-ahead outlook for the economy, 61% reported that they expected bad times ahead.

Sentiment among Democrats continued to slide, back to the lowest since Dec 2020 (as Republican confidence rebounds)….

Source: Bloomberg

Interestingly, buying conditions for houses rebounded in early January but for vehicles it crashed to a record low…

Source: Bloomberg

Finally, and perhaps most notably, inflation expectations rose significantly in January, with medium-term expectations rising to 3.1% – the highest since January 2011…

Source: Bloomberg

So, will The Fed’s jawboning bring down inflation expectations?

As UMich noted, while the Delta and Omicron variants certainly contributed to this downward shift, the decline was also due to an escalating inflation rate.

Tyler Durden Fri, 01/14/2022 – 10:07

Author: Tyler Durden

Economics

How Goldman Is Convincing Its Clients Not To Freak Out About Fed Rate Hikes

How Goldman Is Convincing Its Clients Not To Freak Out About Fed Rate Hikes

Rate hikes are now just right around the corner and traders are…

How Goldman Is Convincing Its Clients Not To Freak Out About Fed Rate Hikes

Rate hikes are now just right around the corner and traders are freaking out, but not so fast according to Goldman.

Following the FOMC meeting in mid-December, and especially last week’s FOMC minutes and the subsequent jawboning by various Fed officials,, it has become clear that the Fed will not only double the pace of tapering but also signaled three hikes in 2022. As a result, virtually all sell-side economists – even stern holdouts such as Morgan Stanley and Bank of America – have raised their forecast from three hikes in 2022 to four – with the first hike now expected to occur in March. Their forecast reflects the greater sense of urgency on behalf of FOMC participants towards quelling inflation, which rose to a four-decade high of 7% as measured by the latest year/year CPI. Why this urgency? Because as one can imagine, Biden was very clear in what Powell’s mandate was when he was renominated: “crush inflation as it is crushing my approval ratings”, because as BofA’s Michael Hartnett noted on Friday, “US inflation is up from 1.4% to 7.0%, while Biden’s approval rating is down from 56% to 42% past 12 months.”

But why is the Fed rushing to hike when a growing chorus of economists now agrees with us that the Fed is hiking right into a recession (or alternatively, hiking to create a recession) an observation that was validated by Friday’s dismal retail sales data… and even without validation, the endgame is clear: as David Rosenberg noted recently, every time the US has had 5%+ inflation, it ended in recession.

Well, according to Goldman’s David Kostin, the unprecedented strength of the labor market has made the Fed more sensitive to high inflation and less sensitive to slowing growth. Alongside rising inflation, the Fed has also cited strong employment data as a catalyst for earlier liftoff and balance sheet reduction. The unemployment rate now stands at 3.9%, falling slightly below the FOMC’s 4.0% median estimate of its long-term level (although looking ahead, Kostin notes that surveys of workers and businesses indicate wage growth is expected to slow to about 4% this year).

To be sure, the market already reflects this and real and nominal rates have both jumped in anticipation of the upcoming tightening cycle. Since the December FOMC meeting, the 10-year US Treasury yield has surged by 26 bp to 1.77%. Consistent with historical experience, equities have struggled amid this rapid rise in yields, and the fastest-growing and longest-duration pockets of the market – i.e., the biggest bubbles such as profiless tech names, the ARKK ETFs, SPACs and so on – have de-rated most.

As a quick aside, perhaps the main reason for the equity puke in the past two weeks is not so much the jump in absolute yield in the past month, but the speed of the move. As Goldman showed in a separate report earlier this week (also available to professional subs), regardless of the level of interest rates, equities react poorly to sharp changes in the interest rate environment, and the past week has been no exception: “Historically, equity prices have declined when interest rates rose by two standard deviations or more. This is true for both nominal and real interest rates across both weekly and monthly periods. The two standard deviation threshold was exceeded on both horizons last week, and the accompanying equity weakness followed the usual historical pattern.”

But while that may explain short-term moves, surely higher rates will lead to longer-term weakness no matter what. And while the answer is yes, the next table shows the sensitivity of the S&P 500 forward P/E multiple to various interest rate and ERP scenarios. Goldman’s interest rate strategists forecast a continued rise in real interest rates that will lift the nominal 10-year Treasury yield to 2% by year-end 2022 (more below), however they also expect the ERP to compress modestly from current levels as the pandemic recovery continues and economic policy uncertainty surrounding potential reconciliation legislation passes. In this base case scenario, the S&P 500 P/E multiple would remain roughly flat this year, allowing earnings growth to lift the index price level. But, if the ERP were to rise to its 10-year median and the Treasury yield rises to 2.25%, the P/E multiple would compress by roughly 15% to 17x, and not even Goldman can spin that as positive.

In any case, as Kostin writes in his latest Weekly Kickstart, market pricing and client conversations indicate investors are braced for a string of hikes in 2022, and as a result, questions from Goldman clients during the past two weeks “have focused on the relationship between equities and interest rates, indicating that the hawkish FOMC pivot is being actively assessed by equity investors.” Moreover, the overnight index swap (OIS) market is currently pricing 3.6 rate hikes in 2022 and 2.6 in 2023, just below the 4 and 3 hikes, respectively, that Goldman forecasts (spoiler alert: the total number of rate hikes will be far less once stocks crash).

And this is where Goldman enters the bullish spin cycle, because the bank makes much more money when its clients buy (only to sell in the future), than selling now. So to ease client concerns that the bottom is about to fall off the market, Kostin writes that “historically” (because clearly we have had many “historical examples” when the Fed’s balance sheet was 45% of US GDP), the S&P 500 index has been resilient around the start of Fed hiking cycles, noting that “although the index has returned -6% on average during the three months following the first hike of recent cycles, the weakness has been short-lived as returns average +5% during the six months following the first hike.” Moreover, as Goldman shows in exhibit 3, the S&P 500 P/E is typically flat during the 12 months around the first hike.

Drilling down into segments, Goldman notes that cyclical sectors and Value stocks outperform around the first Fed hike. The reason: the start of Fed hiking cycles (usually) tends to coincide with a strong economy, which can help to lift cyclical sectors (Materials, Industrials, Energy). However, this time around it is starting as the economy is rapidly slowing yet inflation remains stubborn due to supply-chain blockages, and as such anything Goldman suggests you should do, please ignore it.

Which is probably also true for factors. According to Kostin, at the factor level, Value stocks tend to outperform in the months before and after the first hike: “High quality factors (e.g., high margins, strong balance sheets) underperform in the strong economic environment preceding hikes and outperform in the months following the initial rate increase. Growth is the worst performing factor in the 6 months around the first hike.” Here too, we would flip this 180 degrees because the Fed is now hiking to effectively start a recession (or as the US is already en route to one), so what one should be selling is value while buying growth ahead of the next rate cuts/QE which are now just a matter of time.

Next, Kostin brings out the heavy artillery and urges his skittish clients to consider that “surprisingly” equities have historically performed well alongside rising expectations for Fed hikes. Here, the bank examines the six-month periods since 2004 when OIS pricing of the 5-year-ahead fed funds rate increased by 25 bps, excluding episodes when the Fed was cutting rates.

During these episodes, nominal 10Y yields typically rose by 52 bps with roughly even contributions from real yields and breakevens. Despite this, the S&P 500 returned 9% (vs. its unconditional 6-month average of 5%). Higher earnings expectations drove these rallies as increases in fed funds pricing usually coincided with improving expectations for economic growth. However, as we have repeatedly warned and as even Kostin concedes, “the current inflation-led hiking cycle may prove more challenging for equities.” We are not sure this will boost the confidence level of Goldman clients who are on the fence to just BTFD…

After the initial stage, when markets price more eventual rate hikes, cyclical sectors typically outperform while bond proxies lagged according to Goldman. Industrials, Consumer Discretionary, and Materials outperform the S&P 500 on average during these episodes, with financials especially sensitive to the long-term interest rate outlook and also outperforming. Meanwhile, bond proxy sectors such as Utilities and Consumer Staples underperformed sharply.

As noted above, value has typically outperformed alongside rising market expectations for Fed hiking, but only in cases when the the hiking cycle was led by growth expectations, not to crush inflation, so this time one can argue that everything will be flipped. And while traditionally, small-caps also outperformed, as “quality” factors underperformed, the recent weakness in small-caps confirms that this is anything but an ordinary rate hike cycle.

Curiously, even in his bullish pitch to clients, Kostin – perhaps hoping to preserve some credibility- admits that this is not a typical rate hike cycle, and the recent hawkish pivot has been driven not by “improving growth expectations but by inflation risks” yet even so Goldman’s economists expect growth to remain above-trend in 2022 because, of course, what else can they do: start sounding like Zero Hedge and admit that the Fed is hiking into a recession.

And indeed, Kostin admits that “fading expectations for fiscal stimulus and the hit from Omicron have led our economists to downgrade their growth outlook in recent weeks” however – perhaps unwilling to piss off Biden too much – they still forecast 3.4% GDP growth this year, a stepdown from the 5-6% pace in 2021 but still above their 1.75% estimate of trend growth. Translation: the US will be in a recession by the midterms, courtesy of the Fed.

So after all that, if Goldman clients aren’t running for the hills, maybe the will BTFD after all, and for them, Kostin writes that investors “should balance their exposures to Growth and Value” as Goldman’s rates strategists expect yields will continue to rise, a dynamic that should support Value over Growth, unless of course we enter stagflation at which point all is lost (incidentally, as noted last week, Goldman expects nominal 10-year yield to hit 2.0% by year-end 2022 (with real rates rising to -0.70% almost where they are now) and 2.3% by the end of 2023).

From a growth perspective, Goldman economists expect the waning of the Omicron wave to lift GDP growth from 2% in 1Q to 3% in 2Q, supporting Value stocks. But they expect growth will slow to a 2% pace by 4Q 2022, the type of environment that generally supports Growth stocks. Translation: yes, growth stocks are getting crushed now, but as soon as the current whisper of a recession/stagflation becomes a chorus, watch as “growth stocks” (i.e., the bubble/bitcoin baskets) explode higher and surpass their previous all-time highs.

In short, Goldman’s current recommended sector overweights reflect a barbell of Growth and Value:

  • Info Tech remains the bank’s long-standing overweight due to its secular growth and strong profit margins.
  • Financials should benefit from rising interest rates
  • Health Care combines secular growth qualities with a deep relative valuation discount.

Finally, from a thematic perspective, Goldman continues to recommend investors own highly profitable growth stocks relative to growth stocks with low or no profitability. To this, all we can add is that with low growth stocks having been absolutely nuked by now, the highest convexity when the recessionary turn comes, will be precisely in the no profitability growth sector, which will double in no time once the coming recession/easing cycle becomes the dominant narrative.

Tyler Durden
Sat, 01/15/2022 – 19:00







Author: Tyler Durden

Continue Reading

Economics

Labor Shortage And Surging Shipping Costs Are Biggest Drivers Of US Food Inflation

Labor Shortage And Surging Shipping Costs Are Biggest Drivers Of US Food Inflation

Setting aside the ever-present issue of the global supply…

Labor Shortage And Surging Shipping Costs Are Biggest Drivers Of US Food Inflation

Setting aside the ever-present issue of the global supply chain crunch presently gestating in the PROC, where factories and ports are struggling with the most restrictive lockdown measures since the (Fauci-funded) “China virus” first burst forth out of Wuhan, the US is still facing serious shortages of workers and critical goods like foodstuffs and medicine.

The US labor market disappointed once again in December, while November’s similarly disappointing number was revised up only slightly. Meanwhile, those who are working are struggling with the fact that inflationary price pressures are hammering real wages. And regardless of what the Fed does next, it appears kinks in the economy created by COVID and the federal government’s response to COVID will continue to push food prices higher for the foreseeable future, as Reuters reports.

Citing three critical factors, high demand for groceries, soaring freight costs and “omicron-related” labor shortages, Reuters projects that prices for “food and fresh produce” will continue to climb for the foreseeable future.

Already, growers across the West and Midwest are paying 3x the freight costs from before the pandemic – all to guarantee shipment of perishables like berries and lettuce before they spoil.

Some companies are even holding back on shipping certain goods (like long-lasting onions) to see if shipping costs might ease.

Shay Myers, CEO of Owyhee Produce, which grows onions, watermelons and asparagus along the border of Idaho and Oregon, said he has been holding off shipping onions to retail distributors until freight costs go down.

Myers said transportation disruptions in the last three weeks, caused by a lack of truck drivers and recent highway-blocking storms, have led to a doubling of freight costs for fruit and vegetable producers, on top of already-elevated pandemic prices. “We typically will ship, East Coast to West Coast – we used to do it for about $7,000,” he said. “Today it’s somewhere between $18,000 and $22,000.”

One CongAgra subsidiary blamed labor shortages for the bulk of their troubles.

Birds Eye frozen vegetables maker Conagra Brands’ CEO Sean Connolly told investors last week that supplies from its U.S. plants could be constrained for at least the next month due to Omicron-related absences.

Earlier this week, Albertsons CEO Vivek Sankaran said he expects the supermarket chain to confront more supply chain challenges over the next four to six weeks as Omicron has put a dent in its efforts to plug supply chain gaps.

The packaged goods industry is missing more than 100K workers. Participants filled measly 1,500 jobs last month, according to the BLS data.

The situation is not expected to abate for at least a few more weeks, Katie Denis, vice president of communications and research at the Consumer Brands Association said, blaming the shortages on a scarcity of labor. The consumer-packaged goods industry is missing around 120,000 workers out of which only 1,500 jobs were added last month, she said, while the National Grocer’s Association said that many of its grocery store members were operating with less than 50% of their workforce capacity.

Of course, labor shortages aren’t the only issue. Demand for groceries remains sky high as millions of Americans remain hunkered down, too scared to eat at a restaurant – or too tapped out from their financial difficulties to justify dining at one.

Tyler Durden
Sun, 01/16/2022 – 08:45


Author: Tyler Durden

Continue Reading

Economics

Europe’s Spendthrifts Are Stuck In Irreversible Debt-Traps

Europe’s Spendthrifts Are Stuck In Irreversible Debt-Traps

Authored by Alasdair Macleod via GoldMoney.com,

A Euro Catastrophe Could Collapse…

Europe’s Spendthrifts Are Stuck In Irreversible Debt-Traps

Authored by Alasdair Macleod via GoldMoney.com,

A Euro Catastrophe Could Collapse It

This article looks at the situation in the euro system in the context of rising interest rates. Central to the problem is role of the ECB, which through monetary inflation embarked on a policy of transferring wealth from fiscally responsible member states to the spendthrift PIGS and France. The consequences of these policies are that the spendthrifts are now ensnared in irreversible debt traps.

Even in a Keynesian context, the ECB’s monetary policy is no longer to stimulate the economy but to keep the spendthrifts afloat. The situation has deteriorated so that Eurozone commercial banks appear to have credit restricted in New York, evidenced by the reluctance of the US banks to enter into repo transactions with them, leading to the market failure in September 2019 when the Fed had to intervene.

An examination of the numbers strongly suggests that even Eurozone banks, insurance companies and pension funds are no longer net buyers of Eurozone government debt. It could be because the terms are unattractive. But if that is the case it is an indictment of the ECB’s asset purchase programmes deliberately suppressing rates to the point where they are unattractive, even to normally compliant investors.

Consequently, without any savings offsets, the ECB has gone full Rudolf Havenstein, and is following similar inflationary policies to those that impoverished Germany’s middle classes and starved its labourers and the elderly in 1920-1923. That the German people are tolerating such an obvious destruction of their currency for the third time in a hundred years is simply astounding.

Institutionalised Madoff

Schemes to pilfer from people without their knowledge always end in disaster for the perpetrators. Central banks using their currency seigniorage are no exception. But instead of covering it up like an institutionalised Madoff they use questionable science to justify their openly fraudulent behaviour. The paradox of thrift is such an example, where penalising savers by suppressing interest rates supposedly for the wider economic benefit conveniently ignores the theft involved. If you can change the way people perceive reality, you can get away with an awful lot.

The mass discovery by the people of the fraud perpetrated on the people by those supposedly representing the people is always the reason behind a cycle of crises and wars. It can take a long period of suffering before an otherwise supine population refuses to continue submitting unquestionably to authority. But the longer the condition exists, the more oppressive the methods that the state uses to defer the inevitable crisis become. Until something finally gives. In the case of the euro, we have seen the system give savers no interest since 2012, while the quantity of money and credit in circulation has debased it by 63% (measured by M3 euro money supply).

Furthermore, prices can be rigged to create an illusion of price stability. The US Fed increased its buying of inflation-linked Treasury bonds (TIPS) since March 2020 at a faster pace than they were issued by the US Treasury, artificially pushing TIPS prices up and creating an illusion that the market is unconcerned about price inflation.

But that is not all. Government statisticians are not above fiddling the figures or presenting figures out of context. We believe the CPI inflation figures are a true reflection of the cost of living, despite the changes over time in the way prices are input. We believe that GDP is economic growth — a questionable concept — and not growth in the quantity of money. We even believe that monetary inflation has nothing to do with prices. Statistics are designed to deceive. As Lord Canning said 200 years ago, “I can prove anything with statistics but the truth”. And that was before computers, which have facilitated an explosion in the quantity of questionable statistics. Can’t work something out? Just look at the stats.

A further difference between Madoff and the state is that the state forces everyone to submit to its monetary frauds by law. And since as law-abiding citizens we respect the law, we even despise those with the temerity to question it. But in the process, we hand enormous power to the monetary authorities, so should not be surprised when that power is abused, as is the case with interest rates and the dilution of the state’s currency. And it follows that the deeper the currency fraud, when something gives, the greater is the ensuing crisis.

The best measure of market distortions from deliberate actions of the monetary authorities we have is the difference between actual bond yields and an estimate of what they should be. In other words, assessments of the height of negative real yields. But any such assessment is inherently subjective, with markets and statistics either distorted, rigged, or unable to provide the relevant yardstick. But it makes sense to assume that the price impact, that is the adjustment to bond prices as markets normalise, is greatest for those where nominal bond yields are negative. This means our focus should be directed accordingly. And the major jurisdictions where this applies is Japan and the Eurozone.

The eurozone’s banking instability

A critique of Japan’s monetary policy must be reserved for a later date, in order to concentrate on monetary and economic conditions in the Eurozone. The ECB first reduced its deposit rate to 0% in July 2012. That was followed by its initial introduction of negative deposit rates of -0.1% in June 2014, followed by -0.2% later that year, -0.3% in 2014, -0.4% in 2016 and finally -0.5% in September 2019. The last move coincided with the repo market blow-up in New York, the day that the transfer of Deutsche Bank’s prime dealership to the Paris based BNP was completed.

We can assume with reasonable certainty that the coincidence of these events showed a reluctance of major US banks to take on either of these banks as repo counterparties, as hedge and money funds with accounts at Deutsche decided to move their accounts elsewhere, which would have blown substantial holes in Deutsche’s and possibly BNP’s balance sheets as well, thereby requiring repo cover. The reluctance of American banks to get involved would have been a strong signal of their reluctance to increasing their counterparty exposure to Eurozone banks.

We cannot know this for sure, but it is the logical explanation for what happened. In which case, the repo crisis in New York was an important advance warning of the fragility of the Eurozone’s monetary and banking system. A look at the condition of the major Eurozone global systemically important banks (G-SIBs) in Table A, explains why.

Balance sheet gearing for these banks is roughly double that of the major US banks, and except for Ing Group, deep price-to-book discounts indicate a market assessment of these banks’ credit risk as exceptionally high. Other Eurozone banks with international counterparty business deemed not significant enough to be labelled as G-SIBs but still capable of transmitting systemic risk could be even more highly geared. The reasons for US banks to limit their exposure to the Eurozone banking system on these grounds alone are compelling. And the persistence of price inflation today is a subsequent development, likely to expose these banks as being riskier still because of higher interest rates on their exposure to Eurozone government and commercial bonds, and defaulting borrowers.

The euro credit cycle has been suspended

When banks buy government paper, it is usually because they see it as the risk-free alternative to expanding credit to non-financial private sector actors. In the normal course of an economic cycle, it is inherently cyclical. Both Basel and national regulations enhance the concept that government debt is risk-free, giving it a safe-haven status in times of heightened risk. In a normal bank credit cycle, banks will tend to hold government bills and bonds with less than one year’s maturity and depending on the yield curve will venture out along the curve to five years at most.

These positions are subsequently wound down when the banks become more confident of lending conditions to non-financial borrowers when the economy improves. But when economic conditions become stagnant and the credit cycle is suspended due to lack of recovery, banks can accumulate positions with longer maturities.

Other than the lack of alternative uses of bank credit, this is for a variety of reasons. Trading desks increasingly seek the greater price volatility in longer maturities, central banks encourage increased commercial bank participation in government bond markets, and yield curve permitting, generally longer maturities offer better yields. The more time that elapses between investing in government paper and favouring credit expansion in favour of private sector borrowers, the greater this mission creep becomes.

As we have seen above, the ECB introduced zero deposit rates nearly 10 years ago, and private sector conditions have not generated much in the way of bank credit funding. Lending from all sources including securitisations and bank credit to a) households and b) non-financial corporations since 2008 are shown in Figure 1.

Before the Covid pandemic, total lending to households had declined from $9 trillion equivalent in 2008 to $7.4 trillion in 2019 Q4. And for non-financial corporations, total lending declined marginally over the same period as well. Admittedly, this period included a credit slump and recovery, but on a net basis lending conditions stagnated.

But bank credit for these two sectors will have contracted, allowing for net bond issuance of collateralised consumer debt and by corporations securing cheap finance by issuing corporate bonds at near zero interest rates, which are contained in Figure 1.

Following the start of the pandemic, lending conditions expanded under government direction and borrowing by both sectors increased substantially.

Meanwhile, over the same period bond issuance to governments increased, particularly since the pandemic started, illustrated in Figure 2.

The charts in Figures 1 and 2 support the thesis that credit expansion and bond finance had, until recently, disadvantaged the non-financial private sector. The expansion of government borrowing has been entirely through bonds bought by the ECB, as will be demonstrated when we look at the euro system balance sheet. They confirm that zero and negative rates have not stimulated the Eurozone’s economies as Keynesians theorised. And the increased credit during the pandemic reflects financial support and not a renewed attempt at Keynesian stimulation.

The purpose of debt expansion is important because the moment the supposed stimulus wears off or interest rates rise, we will see bank credit for households and businesses begin to contract again. Only this time, there will be a heightened risk for banks of collateral failure. And higher interest rates will also undermine mark-to-market values for government and corporate bonds on their balance sheets, which could rapidly erode the capital of Eurozone banks, given their exceptionally high gearing shown in Table A above.

Figure 3 charts the euro system’s combined balance sheet since August 2008, the month Lehman failed, when it stood at €1.43 trillion. Greece’s financial crisis ran from 2012-2014, during which time the balance sheet expanded to €3.09 trillion, before partially normalising to €2.01 trillion. In January 2015, the ECB launched its expanded asset purchase programme (APP — otherwise referred to as quantitative easing) to prevent price inflation remaining too low for a prolonged period. The fear was Keynesian deflation, with the HICP measure of price inflation falling to -0.5% at that time, despite the ECB’s deposit rate having been already reduced to -0.2% the previous September.

Between March 2015 and September 2016, the combined purchases by the ECB of public and private sector securities amounted to €1.14 trillion, corresponding to 11.3% of euro area nominal GDP. The APP was “recalibrated” in December 2015, extended to March 2017 and beyond, if necessary, at €60bn monthly. And the deposit rate was lowered to -0.3%. Not even that was enough, with a further recalibration to €80bn monthly in March 2016, with it intended to be extended to the end of the year when it would be resumed at the previous rate of €60bn per month.

The expansion of the ECB’s balance sheet led to the rate of price inflation recovering to 1% in 2017, as one would expect. With the expansion of credit for the non-financial private sector going nowhere (Figures 1 and 2 above), the Keynesian stimulus simply failed in this objective. But when in March 2020 the US Fed reduced its funds rate to 0% and announced QE of $120bn monthly, the ECB did what it had learned to do when in a monetary hole: continue digging even faster. March 2020 saw the ECB increase purchases under the asset purchase programme (APP) and adopt a new programme, the pandemic emergency purchase programme (PEPP). These measures are the reason why the volumes of the Eurosystem’s monthly monetary policy net purchases are higher than ever before, driving its balance sheet total to over €8.5 trillion today.

The ECB’s bond purchases closely matched the funding requirements of national central banks, both being €4 trillion between January 2015 and June 2021. The counterpart to these purchases is an increase in the amount of circulating cash. In other words, the ECB has gone full Rudolf Havenstein. There is no difference in the ECB’s objectives compared with those of Havenstein when he was President of the Reichsbank following the First World War; a monetary policy that impoverished Germany’s middle classes and pushed the labouring class and elderly into starvation by collapsing the paper-mark. Except that today, German society is paying through the destruction of its savings for the spendthrift behaviour of its Eurozone partners rather than that of its own government.

The ECB now has an additional problem with price inflation picking up globally. Producer input prices in Europe are rising strongly with the overall Eurozone HICP rate for November at 4.9% annualised, and doubtless with more rises to come. Oil prices have risen over 50% in a year, and natural gas over 60%, the latter even more on European markets due to a supply crisis of its governments’ own making.

Increasingly, the policy purpose of the ECB is no longer to stimulate the economy, but to ensure that spendthrift member state deficits are financed as cheaply as possible. But how can it do that when on the back of soaring consumer prices, interest rates are now going to rise? Clearly, the higher interest rates go, the faster the ECB will increase its balance sheet because it is committed to not just covering every Eurozone member state’s budget deficit but the interest on their borrowings as well.

But there’s more. In a speech on 12 October, Christine Lagarde, the President of the ECB indicated that it stands ready to contribute to financing the transition to carbon neutral. And in a joint letter to the FT, the President of France and Italy’s Prime Minister called for a relaxation of the EU’s fiscal rules so that they could spend more on key investments. This is a flavour of what they said:

“Just as the rules could not be allowed to stand in the way of our response to the pandemic, so they should not prevent us from making all necessary investments,” the two leaders wrote, while noting that “debt raised to finance such investments, which undeniably benefit the welfare of future generations and long-term growth, should be favoured by the fiscal rules, given that public spending of this sort actually contributes to debt sustainability over the long run.”

The rules under the Stability and Growth Pact have in fact been suspended, and are planned to be reapplied in 2023, But clearly, these two high spenders feel boxed in. The Stability and Growth Pact will almost certainly be eased — being a charade, rather like the US’s debt ceiling. The trouble is Eurozone governments are too accustomed to inflationary finance to abandon it.

If the ECB could inflate the currency without the consequences being apparent, there would be no problem. But with prices soaring above the mandated 2% target that is no longer true. Up to now, the ECB has been in denial, claiming that price pressures will subside. But we know, or should know, that a rise in the general level of prices is due to monetary expansion, the excessive plucking of leaves from the magic money tree, particularly at an enhanced rate since March 2020 which is yet to be reflected fully at the consumer level. And in its duty to fund the PIGS government deficits, the ECB’s balance sheet expansion through bond purchases is sure to continue.

Furthermore, if bond yields do rise, it will threaten to undermine the balance sheets of the highly geared commercial banks.

The commercial banks position

With the economies of Eurozone member states stifled by the ECB’s management of monetary affairs since the Lehman crisis in 2008 and by more recent covid lockdowns, the accumulation of bad debts at the commercial banks is a growing threat to the entire financial system. Table A above, of the Eurozone G-SIBs’ operational gearing and their share ratings, gives testament to the problem.

So far, bad debts in Italian and other PIGS banks have been reduced, not by their being resolved, but by them being used as collateral for loans from national central banks. Local bank regulators deem non-performing loans to be performing so they can be hidden from sight in the ECB’s TARGET2 settlement system. Together with the ECB’s asset purchases conducted through national central banks, these probably account for most of the imbalances in the TARGET2 cross-border settlement system, which in theory should not exist.

The position to last October is shown in Figure 4. Liabilities owed to the Bundesbank are increasing again at record levels, while the amounts owed by the Italian and Spanish central banks are also increasing. These balances were before global pressures for rising interest rates materialised. Given the sharp increase in bank lending to households and non-financial corporations since March last year (see Figure 1), bad debts seem certain to accumulate at the banks in the coming months. This is likely to undermine collateral values in Europe’s repo markets, which are mostly conducted in euros and almost certainly exceed €10 trillion, having been recorded at €8.3 trillion at end-2019.[vi] The extent to which national central banks have taken in repo collateral themselves will then become a major problem.

It is against the background of negative Euribor rates that the repo market has grown. It is not clear what role negative rates plays in this growth. While one can see a reason for a bank to borrow at sub-zero rates, it is harder to justify lending at them. And in a repo, the collateral is returned on a pre-agreed basis, so it’s removal from a bank’s books is temporary. Nonetheless, this market has grown to be an integral part of daily transactions between European banks.

The variations in collateral quality are shown in Figure 5. This differs materially from repo markets in the US, which is almost exclusively for short-term liquidity purposes and uses high quality collateral only (US Treasury bills and bonds and agency debt).

Bonds rated BBB and worse made up 27.7% of the total collateral in December 2019. In Europe and particularly the Eurozone rising interest rates can be expected to undermine collateral ratings, which with increasing Euribor rates will almost certainly contract the size of the market. This heightens the risk of a liquidity-driven systemic failure, as repo liquidity is withdrawn from banks that depend upon it.

Government finances are out of control

The first column in Table B shows government debt to GDP, which is the conventional yardstick of government debt measurement relative to the economy. The second column shows the proportion of government spending in the total economy relative to GDP, enabling us to derive the third column. The base for government revenue upon which paying down its debt ultimately rests is the private sector, and the third column shows the extent to which and where this true burden lies.

It exposes the impossible position of countries such as Greece, Italy, France, and Belgium, Portugal and Spain, where, besides their own private sector debt burdens, citizens earning their livings without being paid by their governments are assumed by markets to be responsible for underwriting their governments’ debts.

The hope that these countries can grow their way out of their debt is demolished in the context of the actual tax base. It is now widely recognised that will already high levels of taxation further tax increases will undermine these economies.

We can dismiss as hogwash the alterative, the vain hope that yet more stimulus in the form of a further increase in deficits will generate economic recovery, and that higher tax revenues will follow to normalise public finances. It is a populist argument amongst some free marketeers today, citing Ronald Reagan’s and Margaret Thatcher’s successful economic policies. But in those times, the US and UK governments were not nearly so indebted and their economies were able to respond positively to lower taxes. Furthermore, price inflation was declining then while it is increasing today.

And as a paper by Carmen Reinhart and Ken Rogoff pointed out, a nation whose government debt exceeds 90% of GDP has great difficulty growing its way out of it.[vii]Seven of the Eurozone nations already exceed this 90% Rubicon, and their debts are still growing considerably faster than their GDP. At 111% the entire Euro area itself is well above it. Taking account of the smaller proportion of private sector activity relative to those of their governments highlights the difference between the current situation and that of nations that managed to pay down even higher debt levels after the Second World War by gently inflating their way out of a debt trap while their economies progressed in the post-war environment.

Additionally, we should bear in mind future government liabilities, whose net present values are considerably greater than their current debt. Over time, these must be financed. And with rising price inflation, hard costs such as healthcare escalate them even further. The position gets progressively worse as these mandated costs become realised.

There is a solution to it, and that is to cut government spending so that its budget always balances. But for socialising politicians, slashing departmental budgets is the equivalent of eating their own children. It is a reversal of everything they stand for. And it requires welfare legislation to be rescinded to stop the accumulation of future welfare costs. There is no democratic mandate for that.

Conclusion

Rising interest rates globally will affect all major currencies, and for some of them expose systemic risks. An examination of the existing situation and how higher interest rates will affect it points to the Eurozone as being the most likely global weak spot.

The Eurozone’s debt position pitches the entire global financial and economic system further towards a debt crisis than generally realised. Particularly for Greece, Italy, France, Belgium, Portugal, and Spain in that order of indebtedness, the problem is most acute. They only survive because the ECB ensures they can pay their bills by funding them totally through inflation of the quantity of euros in circulation. The ECB’s entire purpose has become to transfer wealth from the more fiscally prudent member states to the spendthrifts by debasing the currency.

In the process, based on figures provided by the Bank for International Settlements the banking system is contracting credit to the private sector, and it is not even accumulating government bonds, which is a surprise.  Much like banks in the US, Eurozone banks have become increasingly distracted into financial activities and speculation. The difference is the high level of operational gearing, up to thirty times in the case of one major French bank, while most of the US’s G-SIBs are geared about 11 times on average.

This article points to these disparities between US and EU banking risks having been a factor in the US repo market failure in September 2019. And we can assume that the Americans remain wary of counterparty exposure to Eurozone banks to this day.

That the ECB is funding net government borrowing in its entirety indicates that even investing institutions such as pension funds and insurance companies, along with the banks are sitting on their hands with respect to government debt. It means that savings are not offsetting the inflationary effects of government bond issues. It represents a vote to stay out of what has become a highly troubling and inflationary situation. The question arises as to how long this extraordinary situation can continue.

It must come to an end some time, and by destabilising a highly leveraged banking system the end will be a crisis. With its GDP being similar in size to China’s (which is seeing a more traditional property crisis unfolding at the same time) a banking crisis in the Eurozone could be the trigger for dominoes falling everywhere.

As for the euro’s future, it seems unlikely that the ECB has the capability of dealing with the crisis that will unfold. It has cheated the northern states, particularly Germany, the Netherlands, Finland, Ireland, the Czech Republic, and Luxembourg to the benefit of spendthrifts, particularly the political heavyweights of France, Italy and Spain. It is a rift likely to end the euro system and the ECB itself. The deconstruction of this shabby arrangement should prove the end of the euro and possibly of the European Union itself.

Tyler Durden
Sun, 01/16/2022 – 07:00











Author: Tyler Durden

Continue Reading

Trending