In 2009, the Federal Reserve's balance sheet repaired the high-yield bond market while the labor market was still collapsing. That order of events is the whole mechanism. The Fed did not lend into the economy. It took safe, long-dated bonds away from investors who needed returns, and those investors bought the next-riskiest thing they could find.
March 2009: credit healed while unemployment rose
On March 18, 2009, the Federal Open Market Committee announced it would buy up to $300 billion of longer-term Treasuries. It also raised its mortgage-backed securities program by $750 billion. The S&P 500 had closed at 676.53 nine days earlier, its low of the cycle. High-yield credit spreads had peaked above 20 percentage points in December 2008. A credit spread is the extra yield a risky borrower pays over a Treasury of similar maturity.
By the end of 2009, high-yield spreads had narrowed to roughly 6 points. The S&P 500 had risen more than 60% from its low. US unemployment did not peak until October 2009, at 10.0%.
Risk assets priced a full recovery months before the economy showed one. The usual story is that markets "look ahead." That is true but incomplete. Something mechanical was pushing capital toward risk, and that push did not depend on anyone's forecast for jobs.
For you, the lesson is about timing. Balance sheet policy shows up in asset prices quarters before it shows up in economic data, so waiting for macro confirmation means missing the transmission entirely.
Why the Fed buys duration rather than money
Why did a bond purchase move junk bonds and stocks? The Fed was buying duration, and duration is a quantity investors have to hold somewhere.
Duration measures how sensitive a bond's price is to interest rates. A bond with 8-year duration loses about 8% of its value for each 1 percentage point rise in yields. When the Fed buys long-dated Treasuries, it removes that rate risk from private portfolios and holds it itself. The seller gets reserves in return. These are deposits that banks keep at the Fed, and they pay a short-term rate with almost no duration.
The common belief is that QE "prints money" that floods into the economy. Reserves mostly stay inside the banking system. Banks can't lend them out to households in any direct sense. What does change is the composition of private portfolios: less long-dated, rate-sensitive paper, more short-dated cash.
Fewer long bonds are left for private buyers, so their prices rise and their yields fall. Most of that fall comes through the term premium, the extra yield investors demand for locking money up in long bonds instead of rolling short bills. When the Fed absorbs a large slice of duration supply, investors who still want duration compete for a smaller pool, and the term premium compresses.
The practical point: track how much duration the private market has to absorb, not the headline size of the balance sheet. A trillion dollars of bill purchases removes very little duration. A trillion dollars of 10-year notes removes a great deal.
The portfolio rebalancing channel and who does the rebalancing
The portfolio rebalancing channel is the step that carries the policy from Treasuries into risk assets. It works through actors with fixed obligations.
Consider a pension fund that assumes 6.5% annual returns to meet future payouts. Consider an insurer that has promised policyholders a guaranteed rate. Consider a bond fund benchmarked to a total-return target. None of them can accept a lower yield and simply wait. Their liabilities are contractual. When the Fed bids Treasury yields down, these holders sell their Treasuries to the Fed at a good price. Then they face a reinvestment problem: the cash they received earns almost nothing, and the bonds that used to meet their target no longer yield enough.
So they move one step out the risk curve. Treasury holders buy agency mortgage bonds. Mortgage holders buy investment-grade corporates. Corporate holders buy high yield. Equity allocators see lower bond yields and accept lower earnings yields on stocks. Every handoff compresses the risk premium in the next asset class. This is why the effect builds in sequence and does not hit every asset at once.
Economists describe this with preferred habitat theory. Many investors have natural maturity and risk segments they prefer, and they move between segments only when the price difference becomes large enough. QE widens that difference on purpose. The Fed is not trying to persuade these institutions. It is changing the relative prices they face and letting their mandates do the rest.
Corporate borrowers then complete the loop. Lower spreads let companies refinance, extend maturities and avoid default. Fewer defaults justify the narrower spreads that brought them about. In 2009, this refinancing window explains much of why high-yield credit recovered faster than default forecasts predicted.
If you want to know where QE will show up first, look at the assets owned by investors with hard return targets. Credit spreads usually move before equities, and equities move before employment.
A worked example: multiples rise without any change in earnings
The equity side of the transmission can be shown with simple arithmetic. Treat the earnings yield (earnings divided by price, the inverse of the P/E ratio) as the 10-year Treasury yield plus an equity risk premium, the extra return investors demand for owning stocks.
Required earnings yield = 10Y Treasury yield + equity risk premium
P/E = 1 / required earnings yield
Assume an equity risk premium of 4.0% and a 10-year yield of 3.5%. The required earnings yield is 7.5%, so the P/E is 1 / 0.075 = 13.3.
Now say balance sheet purchases compress the term premium and pull the 10-year yield down to 2.5%, with the risk premium unchanged. The required earnings yield falls to 6.5%. The P/E becomes 1 / 0.065 = 15.4.
That is a 15.8% rise in the market multiple with no change at all in corporate earnings. No household spent an extra dollar. No company hired anyone. The rebalancing channel can deliver the equity gain before the economy moves.
The same arithmetic works in reverse, which matters for today. As of October 2, 2026, the 10-year yield stood at 5.28% and the 2-year at 4.83%, a positive 2s10s spread of 0.45 points. Plug 5.28% into the same framework with a 4.0% risk premium and the required earnings yield is 9.28%, a P/E of about 10.8. Equity markets rarely trade at that multiple, because the risk premium itself moves around. A positive and widening curve means long bonds are once again paying investors to take duration. Every point of that compensation is a point that equity and credit have to compete against.
Your practical takeaway: when you see equity multiples expand while earnings estimates stay flat, ask whether the discount rate did the work. If it did, the gain is on loan from the bond market.
When the transmission breaks
The mechanism is not a law of nature. It depends on conditions that have failed before, and each failure is instructive.
The signal channel can overwhelm the quantity channel. On May 22, 2013, Ben Bernanke told Congress the Fed might slow its purchases. The 10-year yield rose from about 1.6% in early May to roughly 3% by September, even though the Fed had not yet bought a dollar less. Investors repriced the future path of duration supply, and emerging-market debt and high-yield funds saw heavy outflows. Expectations about the balance sheet can move markets before the balance sheet itself moves.
Reserves can become scarce. Shrinking the balance sheet drains reserves. In September 2019, overnight repo rates spiked to around 10% intraday as banks hoarded cash. The Fed had to start buying bills again. Past a certain point, the plumbing becomes the binding constraint, and the policy stops behaving like a smooth duration lever.
Fiscal supply can swamp the Fed. The private market absorbs Treasury issuance minus Fed holdings. If the Treasury issues long-dated debt faster than the Fed buys it, the private sector's duration load grows even during QE. If the Fed simply stops reinvesting while deficits stay large, the load grows faster still. A steepening curve with a long-end yield above 5%, as in October 2026, is consistent with the market demanding more term premium to absorb that supply. We would hedge the exact split between supply and expectations; the direction of the pressure is clear.
The rebalancing can go somewhere unintended. Portfolio managers reach for whatever offers yield, not necessarily what helps the real economy. Lower spreads can fund buybacks and leveraged recapitalizations as easily as new factories. That is one reason risk-asset gains under QE often outran the economic recovery they were supposed to signal.
When one of these conditions is present, balance sheet headlines become a weak guide. Watch the term premium and repo market stress before trusting the old relationship.
Reading the present through the same chain
The chain runs in both directions. QE removes duration and pushes holders into credit and equities. Runoff, or heavy issuance with a passive Fed, does the opposite: it returns duration to private hands, rebuilds the term premium, and raises the yield that every risky asset has to beat.
With the 10-year at 5.28%, the first asset to feel that pressure should be the one that felt QE first: high-yield credit. If junk spreads stay tight while long yields rise, credit investors are either betting that the term premium will fall back or ignoring the competition from Treasuries. Either way, the margin for error in high-yield is thinner than spread levels alone suggest, because the all-in yield is doing more of the work.
The mechanism in one sentence: the Fed's balance sheet moves risk assets by changing how much duration private investors must hold, and investors with fixed return targets carry that change from Treasuries into credit and equities before the economy registers anything. If you track the term premium and the private sector's duration load, you will see the transmission coming. If you track only payrolls and the balance sheet headline, you will see it last.
