The Fed does NOT control rates, inflation, or money: They do “expectations”
With 1929-like refrains all too common among Wall Street and the media today — to “not fight the Fed” — everyone seems to know that the Fed is “tightening” and sending the economy into recession, and yet almost no one seems to know what the Fed is doing.
What are open market operations? What rates can the Fed actually set, and what is their market size and relevance (hint: small and very small)? What is QE and what evidence is there that it provides liquidity or “stimulus”?
This post will show that the Fed does not do money, control rates, control inflation, and they told us decades ago that they can’t even track money supply. Instead they follow rates and announce “stimulus” to influence “expectations” (a policy called “suasion”).
The idea that the central bank “controls” the entire economy is based on linear, falsifiable, 100 year old economics and ignores evidence, the nature of complex systems, nonlinearities and the fact that Fed’s participation in the monetary system is a fraction of a fraction of the overall market. The Fed’s rate “targets” correlate neither to inflation, market rates, nor the dollar, nor does “QE”.
So what is QE, what rate does the Fed “control”, what are bank reserves, and how have these policies affected or “stimulated” economies around the world, like Japan?
What is QE (“Quantitative” Easing)?
QE is an asset swap, swapping one type of collateral (i.e. dollar denominated bonds) for another of equal value. As a result, Fed liabilities go up as a simple accounting byproduct.
The important thing to understand about QE is that swapping TBills for MBS, or any bonds/securities, is of equal notional value. Put simply, there is zero new notional value of dollars in the economy with this swap. None. They buy $100 billion in one bond and provide another.
Yes, they can “create” T-bills, but it is the exact same way that a bank creates a deposit — it’s a liability of an asset. Similarly, when a bank deposits money with the Fed IOBR, it “creates” a bank reserve, just as depositing/swapping money with the Fed in certain QE programs “creates” a treasury bill. This T-bill is simply a transferable and collateralizable asset that can be used in trade (unlike a “bank reserve”, which is a non-transferable deposit because they are only available to prime dealers).
The point being that there are no new notional dollar assets in the economy under QE, they are simply assets swapped with the Fed, which therefore moves liabilities to the Fed instead of elsewhere, which is basic accounting.
The Fed having additional offsetting assets is why they can easily unwind QE and decrease their liabilities instantly. If they had “printed” “money” it would be permanently multiplying in the economy, but instead they are swapping assets.
The Fed has never “printed” money
The Fed doesn’t print money, and they have not. They considered printing money after the crash of 1873 — which, like all financial crises, was a monetary shortage. However, Grant got cold feet and worried about the consequences of inflation and the plan was scrapped.
Fed intervention in the Great Depression and the Great Financial Crisis were asset swaps, “open market operations”, or now called “QE”. Note that the 1930s and years following 2008 were deflationary/disinflationary and very much not inflationary eras.
After multiple rounds of QE, to his amazement and the frustration of congress, Fed Chairman Bernanke saw the rate of inflation drop for the next decade.
Japan had 27 rounds of QE over decades and CPI and rates have always gone lower over the long term. Japan is now going into a 4th decade of depression, and their GDP is lower today than 1995, and we are following the exact same “stimulus” (it’s not) policy. (Note that they could use too much debt and eventually risk default, which is inflationary — i.e. “Triffin’s Paradox”).
Many point out that members of the Fed have referred to QE as “money printing” and while this is true, this is not how they discuss it internally. The main point of QE is to provide “confidence” to the markets that something is being done and to increase your inflation expectations.
M1/M2 are useless metrics: money creation is in “shadow banking”
Greenspan said, while head of the Federal Reserve, that “money supply trends veered off path several years ago” and that M1/M2 were now useless metrics and no simple metric could explain something as complex as the American economy. Yet, the media and Wall Street still track and quote these metrics as useful.
The Fed does not create money, commercial bank lending creates money. In the 60s and 70s there was a boom in shadow banking dollar deposits (hence two decades of inflation) and the repo system over time evolved from a “medium of exchange” into shadow banking today.
To say the vast majority of dollar lending is in the shadows would be quite an understatement. It dwarfs all money supply metrics but because it is offshore, off balance sheet, a result of collateral reuse and derivatives, it is impossible to track. This is why M3 was abandoned, which was supposed to include MMF and the shadows. But how could they track something they don’t even require to be disclosed (which is true to this day)?
Bear Stearns in 2007 and Lehman in 2008 were bank runs in the repo market. Few understand this and the importance of it. Lehman, like LTCM, had multiples of its market cap or assets in repo obligations (nearly a trillion dollars in repo obligations). Repo is the dog that wags the market’s tail and our Fed doesn’t even track it.
Which is to say, when you see charts of M1 and M2 know that you are looking at a fraction of the money supply and the rest is in the shadows (which I will dedicate posts to over the next months). The bank run and credit contraction of 2008 happened in the [undisclosed] repo market and wouldn’t show up on any money supply metric.
Does the Fed cause market tightening?
Before I explain what the Fed Funds Rate actually does, let’s start with the evidence: the Fed does not remotely control rates; they follow rates… and this too is easily verifiable.
Keep in mind that the Fed wants to take credit for “raising rates” to show that they control money… Yet, no one seems to stop and question whether or not the great Wizard is actually in control of Oz.
Exhibit A: Bond yields have been rising since Aug 5, 2020
Rates have been rising since Aug 5, 2020…
And the Fed announced “QT” March 17, 2022!
How is that causation? Did the Fed really cause rates to rise for the year and 7+ months before raising rates? Remember, they were in QE in 2020 and the Fed was buying bonds to make rates go down, and they went up! Now they take credit for raising rates even though they went up for 19 months prior to QT? How does this make any sense? It doesn’t if you check, just no one questions the great Wizard or Oz.
Exhibit B: Fed’s overnight rate was “raised” to 3%, yet market rate is down to 2.5%.
Now that rates have been going up for the last two years, the great Wizard has jumped in and raised overnight rates to 3.15% (or 3.05% on RRP)… Yet, the 4 week(!) is under 2.5% — 65 bps below the Fed’s overnight rate! Recently it was 80 bps below the Fed overnight rate.
Why would anyone ever pay more for a longer bill with a lower yield? Hard to explain that if you believe in the almighty Wizard controlling “rates”. This view of their “controlling” rates is only possible if you don’t know what it (IORB and now RRP) is, who it is available to, and the relevance to the market (none).
This only makes sense if you realize that the Fed follows rates and does not set them.
Mark my words: treasury yields will follow the yield curve and it will fall, regardless of whether or not the Fed announces “lowering” rates before or after.
Exhibit C: The stock (/everything) bubble started popping long before “QT”
Again, Fed “tightening” was announced on March 17, 2022, yet…
- Nasdaq peaked Nov 19, 2021
- Russell peaked Nov 8, 2021
- S&P peaked Jan 3, 2022
Just like the dot com bubble, the “tech” stocks start to go down at least half a year before the mainstream indices. The stocks that best represent the face of this bubble are best indexed by the various “ARK” funds, which all seemed to peak on Feb 12, 2021:
Now did the Fed cause stocks to peak in early and late 2021 by the “tightening” it announced in spring of 2022… Or maybe, just maybe, are we reliving a bigger version of the dot com bubble and attributing it to the Fed? The evidence strongly indicates that the Fed is following, not leading, the market.
In fact, what evidence is there to the contrary?
I believe shadow banking tightening (actual monetary “tightening”) began in the first half of 2021. Fedwire mysteriously broke in Feb 2021, repo fails started going up, yields started to fall in March until December, and last but not least, reliance on Fed’s RRP:
A chart associated with times of illiquidity and flight to safe collateral began to explode in March 2021 (must be a coincidence). This is a channel for the Fed to provide dollar collateral for those who need to fulfill dollar obligations. We’ll go over this more, but the point here is that it began exploding in early 2021 and has been setting new records.
Yield curve began its path to inversion March 29, 2022:
A lot of monetary coincidences pointing to spring of 2021 and Fed announced tightening a year later.
Exhibit D: There is ZERO correlation between Fed Funds and dollar metrics
Here is the Fed Funds and the DXY:
Same chart from heading into the GFC:
(Quick note, these charts are from Jeff Snider of EuroDollar University. And these slides are from Jeff’s latest interview on MacroVoices podcast, where EuroDollar University originally began — I highly recommend starting from the beginning here!)
As Jeff shows in the details of the chart, there is no correlation between the Fed and the dollar index.
But guess what the DXY does correlate to… Treasury bond prices:
Therefore the Fed Funds Rate neither correlates to the dollar index (DXY) nor dollar bonds (i.e. “rates”, like your mortgage)!
I’m unclear as to the logic of how and why a bank getting higher returns on their cash reserves would possibly translate to how much they would pay for Apple 30 year bonds or charge for mortgages. Is it not obvious that mortgage rates and Apple bond rates go up because of market conditions (like CPI (also not correlated to the Fed)) and not the bank’s savings rate?
Furthermore, is it not common sense that low supply equates to high prices? Why would this be different for bonds? Low rates are associated with tight money, like the 1930s, 2009 and every recession / monetary contraction.
(Note that the DXY often diverges upwards and T-bills follow later — like Sept 29, 2008 when DXY rose and bond prices, then in December both rose. This massive divergence could be a warning.)
Dollar up = tight money
Jeff Snider shows another chart that is correlated with the dollar — the yield curve. Yield curve and the [inverted] DXY:
When the yield curve inverts, that shows tight money, and the DXY goes up.
To further illustrate his point of tight liquidity, here is the inverted DXY and swap spreads. How in the world could derivatives (swaps) of 30 year US treasuries be worth more than the underlying asset? Because there is a general collateral shortage which drives up the price of swaps and the [inverted] DXY:
5 year swaps minus 5 year UST vs [inverted] DXY:
Swaps come with enormous counterparty risk — hence how the financial crisis took off when AIG couldn’t payout on CDS obligations. Yet, they are easier to get your hands on to fulfill obligations (and hedge balance sheets to create more lending capacity).
The dollar market is wagging the economic tail
In early 2020 and 2008 the crisis started with mass dollar liquidations. In 2020 the Fed couldn’t figure out why there was such strong treasury selling (margin collateral demand) and despite Fed buying, prices kept falling. Then suddenly other forms of collateral (corporate bonds and MBS) were no longer accepted in repo and treasury demand soared (flight to safety), dropping 150 basis in two months while other bond yields jumped up.
We are still in this first stage of dollar liquidations and it has lasted for an unusually long time. Treasury prices have dropped (and yields increased) and unusually and suddenly became correlated with the stock market in 2021:
What people don’t understand about the failure of Bear Stearns, Lehman, CDOs, related commercial paper and mortgage securities, is that these assets were used to meet dollar obligations. Almost all major debts around the world are in dollars and those debts create dollar collateral (bonds), but they also create dollar obligations and therefore, like a ponzi, the system increasingly relies on new dollar loans to be created.
What people don’t realize about the fall of Evergrande and mortgage bonds in China, as well as Sovereign bonds around the world, is that this is why Central Banks around the world are liquidating their treasury reserves… but what happens when they run out? Treasuries will spike — just like 2008 and just like 2020.
Don’t attribute the market’s actions of a complex system to the small and linear Fed, (although you can blame lack of regulation to the Fed, since that is their job). Yields are a result of liquidations and if countries and companies run out, bankruptcies and flight to safety begin.
How does the Fed control “rates”?
As shown, the Fed does not control market rates, it follows and the market rates are currently lower than the Fed’s rates… which alone should be dumbfounding, yet I hear no one talking about it.
To understand how this is even possible, we have to understand what instruments the Fed has.
IORB (and IOER) is the Fed’s savings account for primary dealers. These are major US banks and financial institutions that sell securities for the Fed. Instead of the Fed selling to everyone, they sell to a few major firms and then they provide treasury securities to the rest of the market.
One of the benefits of being a prime dealer is that you can deposit money directly with the Fed and therefore all selling and buying (or QE) is just changed on the Fed’s ledger. The Fed Funds Rate is simply changing the interest rate on these deposits at the Fed.
This is considered the only completely safe form of dollar “checking” account (so to speak) deposits, as the dollar bill is already a liability of the Fed and therefore there is no counterparty risk. Otherwise, no financial institution wants to hold “cash”, they want “collateral” (EuroDollars).
Adjusting the annual rate of interest on this savings account no more controls inflation than JP Morgan Chase adjusting their savings rate. In fact, JPM has a similar amount of deposits as the Fed does — one single bank has more dollar liabilities. But of course, neither institution “set” rates; they follow CPI. If this controlled CPI, couldn’t JPM and all the banks together completely control rates, and therefore inflation? By the same [falsifiable] theory, Wall Street has way more control over rates and inflation than the Fed.
Before I jump to RRP, I want to discuss why the idea that the Fed could control the entire market’s rates on all assets with an overnight rate that is a tiny percentage of the market is more absurd than assuming Lehman brothers controlled the entire economy in 2007.
Why is it absurd to think the Fed’s IOER (or RRP) rate controls the economy?
- The Fed is taking deposits from a tiny percentage of the market. In the case of prime dealers, yes, they are important institutions, but they keep vastly more money in collateral in the first place. So you are talking about a tiny percentage of assets of a tiny percentage of institutions globally (when nearly all loans made globally are dollar denominated).
- There is no evidence of long end effects from overnight rates.
- Buying assets, or lending an entirely different asset all together, doesn’t remotely impact the fundamentals that cause defaults and therefore the underlying value of the asset. Which is to say, the Fed buying mortgage backed securities by the billions has zero impact on the American public’s ability to pay back their mortgages (although it profits Wall Street). It is mortgage defaults that drive the price of mortgage assets, not buying and selling. Buying and selling might drive the price of any asset in the short term, but it is cash flows and cash reserves that pay interest that drive value in the long term.
“In the short run, the market is a voting machine but in the long run it is a weighing machine.” — Benjamin Graham - Economies are dynamic complex systems and by their nature can not be controlled.
The example I like to use is a fund that cornered the copper market in the 1700s. They bought up mines and futures and drove up prices while buying all these assets. These higher prices they were now paying to monopolize the industry (to then cut supply and make tons of money,) resulting in the development of a robust copper recycling market, which caused copper prices to collapse, resulting in a historic multi-country European bailout.
Thus, it is naive to believe that a complex system like our economy could be controlled with a tiny overnight rate to a tiny few who keep a tiny portion of their assets in it.
Fed’s ON RRP: What is RRP and why did it boom?
RRP is like IOER, where they pay interest for depositors of the Fed but instead they give collateral (treasury securities), which is more useful. RRP is open to broader use: MMFs, important banks and GSEs globally. You deposit money and the Fed gives you treasury securities. (It states that they do 60 day “term” rates, but it’s all overnight).
In late 2014 we entered what Jeff Snider calls “EuroDollar #3”, referring to the third great dollar collateral shortage (the first being 2007–2009). The FOMC authorized $300 billion to go towards providing collateral.
From reviewing the FOMC transcripts of 2014 it seemed that it was a suggestion by a “participant” (big bank) with the intention of eliminating counterparty risk. Yet, the Fed in all the transcripts I looked at never once mentioned that it would aid during collateral shortages, nor did they ever discuss again that the banks’ viewed this as a safe way (without counterparty risk) to get collateral. Instead, they only ever discussed it as a tool to control rates. It is quite clear that they were still ignorant of collateral shortage issues and viewed it as another “tool” to “intervene” in “normalizing” rates (because they still think they can, that it matters and that it’s “stimulus”).
Collateral is more useful than cash.
Because even cash deposits at banks are uninsured (above $500k I believe), institutions eliminate counterparty risk by having collateral. Because of this and the fact that collateral is now a clearing token (medium of exchange) for all global trade, it is collateral (EuroDollars) that is used by commercial banks, not cash.
Second, treasury securities (and Fed deposits) are considered the least risky because of elimination of counterparty risk and liquidity — if someone fails to unwind an overnight loan (repo), you need to be able to sell it quickly, thus liquidity is top priority in repo. Because treasuries are the deepest and most liquid market in the world, and elimination of counterparty risk, when collateral value is questioned in repo, users are forced to acquire treasuries in volatility.
Third, low yields are a sign of a low supply of collateral. Sure, there is more cash than collateral, but cash does not multiply in our monetary system, collateral does! Bank balance sheet [lending] capacity (Risk Weighted Assets) and commercial bank lending (and reuse) is designed around collateral (EuroDollars), not dollars.
Lastly, it doesn’t make any sense why anyone would lend cash to the Fed at a lower rate when there are lots of forms of higher paying collateral, and this is why historically it only gets used at the end of the quarter (when banks stop rehypothecation/reuse of collateral so it isn’t disclosed in quarterly reports).
In fact, the Fed is not truly involved in repo, and Fed liabilities can not be multiplied (rehypothecation) in commercial bank lending. Thus, as Manmohan Singh of the IMF points out, Fed involvement is inherently a monetary/credit contraction by taking collateral out of reuse.
Nonetheless, RRP is a flight to safety. Knowing that, the RRP chart (shown again here) should be worrying:
RRP continues to set new records and would you describe the last weeks as having “excess liquidity”? RRP, a flight to safety, is a sign of monetary contraction and risk avoidance in the credit markets.
I assume we will soon see what type of collateral (and who) people are fleeing from when we see an increase in spreads between treasuries and other bonds (like 2008 and March 2020). During times of illiquidity there is even a spread between on-the-run treasuries, which are newly issued, and off-the-run treasuries, which I expect we’ll see also when T-bills turn to follow the DXY upward.
Conclusion
There is zero evidence that the Fed controls inflation or rates. They don’t create money, the commercial banks do, and therefore they don’t do “stimulus” or “tightening”.
The important thing to understand is that the Fed does not print money, it only swaps assets via creating liabilities to offset asset deposits. In fact, the EuroDollar market is out of reach of the Fed, being decentralized ledger money of offshore commercial bank entities.
All tools of the treasury and Fed are debt instruments. Neither the Fed liabilities, nor M1/M2, even represent the debt of the government — which is handled by the treasury. Fed liabilities go up and down like the deposits of any other bank and do not represent the money supply.
If you want to understand the amount of debt the government is incurring (actual debt spending), do not look at the Fed’s growth in liabilities, look at the Federal deficit.
That being said, as we see in Japan, the Fed is playing a very dangerous game in order to promote market “stability”. After all, they sold treasuries to buy mortgage backed securities all through 2020. The Fed racked up $2.5 trillion in MBS.
Since then, they lost a lot of taxpayer money:
In other words, the Fed bought $2.5 trillion of MBS via debt, in order to help Wall Street, and they declined ~18%. This had two results:
- It didn’t remotely help the value of mortgage backed securities, which are governed by the economics of people paying back mortgages.
- The Fed bailed out Wall Street and the taxpayers will eventually (or slowly) have to pay the difference.
The Fed “bought the dip” in 2009 and luckily assets went back up (as the EuroDollar system found other [foreign] areas of credit expansion to exhaust), and now like a redditor on WallStreetBets, they think every time you buy the dip it goes back up and the Fed makes money for taxpayers (which happened when they assumed AIG’s assets, including CDS). Ask Japan — it doesn’t always turn out that way. Ask tech stock dip buyers.
The end result, and the nature of debt in general (government and private), is that it frontloads spending. That is true for an individual or a society. However, when a society or government uses enough debt for spending, it can appear inflationary at first, but it creates more obligations (demand) for money later. Therefore debt is deflationary.
Spending on credit, creating offsetting obligations, relies on the exponential growth of debt to create more dollars to have enough currency to make payments on old debts. Since money is a unit of account of resources, when this debt growth exceeds GDP growth (resource growth) it is called a ponzi. There simply will not be enough money to pay obligations without GDP growth exceeding debt growth (there isn’t) and this is deflationary. Debt spending exponentially increases demand for more dollar assets (debt/bonds/EuroDollars) but eventually, in a credit contraction, results in a decreasing money supply.
The Fed is swapping assets like a day-trader (this is what the open market operations desk at the FRBNY does, day-trade) and sometimes they get left with a gain and sometimes they end up with more liabilities than assets. We are entering the phase of deflation and the bag holders are the taxpayers and consumers.
How will this turn out? Japan has had 30 years of deflation and their QE is 12 times ours (relative to GDP), which is a lot of Central Bank debt. I’m not sure if this is at a breaking point of default (which would break their currency) but it’s a lose-lose: default is hyperinflation, and debt multiple times GDP is deflationary.
All monetary crises are deflationary monetary countractions (not enough “medium of exchange” to clear trade and meet obligations, creating a debt spiral). And this gambling with taxpayer money to bailout Wall Street is based on falsifiable economics (regardless, J Powell is a lawyer who knows nothing about economics anyhow) and all evidence over decades and dozens of countries’ QE point to these policies being deflationary.
Most people trust that the Fed is leading us into a recession and it is precisely this blind belief — that the Fed mysteriously controls the economy — that they will rely on when they announce “QE” as “stimulus”… As you will see, it will have no long-term impact on the economy, just Wall Street profits.
