FINRA Margin Debt vs S&P 500 (SPY)

FINRA Rule 4521 requires every registered broker-dealer to report aggregate margin balances on the last business day of each month, and FINRA publishes the consolidated total approximately three weeks later. The series captures every margin loan extended against equity, options, and bond positions held at US broker-dealers and is the most comprehensive single measure of retail and small-institutional leverage available. Advisor Perspectives publishes the cleaned monthly series back to January 1959, which is how analysts compare the current cycle against the 2007 and 2021 cycle peaks.

The historical lead-lag is well-established. Margin debt peaked in July 2007 at approximately $381 billion, three months before the S&P 500 peak in October 2007, and bottomed in February 2009 alongside the equity-market trough that March. Margin debt peaked at approximately $935 billion in October 2021, two months before the SPY peak in early January 2022. The current cycle peaked in January 2026 at $1.279 trillion. By the historical pattern the lead-time would put a SPY peak roughly two to three months after the margin-debt rollover, which means April to May 2026, the same window in which the April 24 S&P 500 print of 7,165.08 was set. The pair is therefore in the highest-information-content phase of its cycle right now. FINRA also publishes the related free-credit-balances series (cash held in margin accounts), which has historically served as an inverse confirmation: free credit balances usually fall as margin debt rises and recover as margin debt unwinds.

The July 2007 peak ran $381 billion (Advisor Perspectives clean series) and represented roughly 2.7 percent of nominal US GDP, a record at that time. SPY peaked October 9, 2007 (24 months before the March 9, 2009 trough), and margin debt fell 54 percent peak-to-trough from $381 billion in July 2007 to $173 billion in February 2009. The drawdown was driven by a combination of forced liquidation through margin calls and voluntary deleveraging. The October 2021 peak ran approximately $935 billion, roughly 3.97 percent of nominal GDP, the highest ratio in the post-WWII record. SPY peaked January 3, 2022 and fell 25 percent peak-to-trough by October 2022. Margin debt fell 36 percent peak-to-trough from $935 billion in October 2021 to $599 billion by September 2022.

The January 2026 peak at $1.279 trillion is roughly 4.0 percent of nominal GDP, narrowly higher than the October 2021 ratio. This ratio matters because the marginal leveraged buyer is the most price-sensitive participant in the equity market, and 4 percent of GDP represents the historical upper bound of sustainable retail leverage. Once that threshold breaks, the deleveraging mechanism amplifies any equity-market drawdown because falling collateral values trigger maintenance-margin calls that force further selling. The 2007 and 2021 cycles both show this asymmetry: margin debt fell 36 to 54 percent against SPY drawdowns of 25 to 50 percent, a beta of approximately 1.5 to 1.

February 2026 margin debt fell 2.0 percent month over month to $1.255 trillion, the first monthly decline in ten months. March fell 2.6 percent to $1.220 trillion, the second consecutive decline. Year over year margin debt remained 38.7 percent higher than March 2025, but the consecutive-month drop is the variable that matters for cycle-position signaling. The Advisor Perspectives March release flagged the two-month sequence as the first sustained rollover since the November 2021 to October 2022 episode that preceded the 2022 bear market.

The interpretation hinges on whether the decline reflects voluntary risk reduction or forced deleveraging. Voluntary declines tend to come in 1 to 3 percent monthly steps over 6 to 9 months and produce shallow equity-market drawdowns of 10 to 15 percent. Forced declines come in 5 to 10 percent monthly steps over 3 to 4 months and produce deep drawdowns of 25 to 50 percent. The current 2.0 to 2.6 percent monthly pattern fits the voluntary template, which is consistent with sophisticated investors trimming exposure ahead of perceived risk events (the April 2026 Iran-related oil shock, the WTI surge above $105) rather than mechanical liquidation. That distinction is the highest-conviction read for the next two quarters. The 2018 December episode is the cleanest historical analog: margin debt fell 4.4 percent in December 2018 as SPY drew down 9 percent, but the decline did not extend into a 2008-style cascade because the December 2018 sell-off resolved at the Powell pivot in early January 2019 rather than producing the kind of multi-month forced-unwind chain that turns a voluntary decline into a forced one.

The 2010 to 2019 era of zero interest-rate policy and ample liquidity produced a sustained margin-debt expansion that ran ahead of fundamental equity-market support. From the February 2009 trough at $173 billion to the December 2019 level of $580 billion the series rose 235 percent against SPY's 309 percent return. Margin debt as a percentage of GDP held in a 2.0 to 2.7 percent band that was structurally higher than the 1.5 percent average of the 1990s, reflecting the lower cost of leverage. The series gave several false signals over this period: short-lived margin-debt declines in 2011 (European debt crisis), 2015 to 2016 (commodity rout), and 2018 (Q4 risk-off) all preceded equity drawdowns of 10 to 15 percent that recovered within months.

The lesson from the 2010s false signals is that the magnitude and duration of the margin-debt rollover matter more than the rollover itself. A 2-month decline of 2 to 3 percent each month, the current April 2026 reading, is the lower bound of what historically counts as a meaningful cycle-position signal. If May and June 2026 produce additional 2 to 5 percent monthly declines, the signal becomes much more robust and the historical base rate for a 15 to 25 percent SPY drawdown within 12 months rises to approximately 65 percent. If May produces a sharp re-acceleration back to a new high, the February to March rollover is more likely to read as a consolidation than as a cycle peak.

FINRA margin debt covers traditional margin lending against securities collateral. It does not capture options leverage, structured-product leverage, or leveraged-ETF AUM. The 2022 to 2026 cycle has seen options leverage grow disproportionately fast: total US equity options volume averaged 47 million contracts per day in 2025 versus 28 million in 2019, with 0DTE volume specifically rising from less than 5 percent of SPX volume in 2019 to 50 percent or more by 2025 according to Cboe data. Leveraged-ETF AUM in single-stock 2x and 3x products reached $50 billion by April 2026 from less than $5 billion in 2019. Both channels add to total retail leverage in ways the FINRA series does not capture.

The practical implication is that the margin-debt-to-GDP ratio at 4.0 percent in January 2026 understates total retail leverage by roughly 50 to 100 percent depending on assumptions about options and leveraged-ETF effective leverage. The composite measure suggests retail leverage is at the highest absolute level in market history, well above the 2021 peak, even as the FINRA series alone is only narrowly higher. Reading the FINRA series in isolation will systematically underestimate the speed and depth of any deleveraging episode this cycle. Cross-checking against Cboe options open interest, leveraged-ETF AUM (TQQQ, SOXL, NVDL specifically), and the SEC short-interest report on a monthly cadence is the disciplined approach.

The closest historical analog to the current configuration is October to November 2021. Margin debt peaked $935 billion in October 2021. SPY made a marginal new high on January 3, 2022. Margin debt had already fallen 6 percent month over month in November 2021, the first material decline. The 25 percent SPY drawdown from January 2022 to October 2022 followed. The current April 2026 setup runs: margin debt peak January 2026 at $1.279 trillion, two consecutive monthly declines totaling 4.5 percent into March, S&P 500 record at 7,165 on April 24. By the 2021 analog the equity peak window is May to June 2026 with a 15 to 25 percent drawdown probability of roughly 50 to 60 percent over the following 12 months.

The operational read is to scale gross equity exposure off the May FINRA release (mid-June 2026). A May print of less than minus 2 percent would extend the consecutive-decline streak to four months, the longest sustained rollover since 2022, and would shift the historical base rate for a 15 to 25 percent SPY drawdown above 65 percent. A May print of plus 2 percent or higher would null the cycle-peak signal and put margin debt back on a path to a new high, which historically extends the equity-market top by another 6 to 9 months. Watch the FINRA release on the 25th of each month for the cleanest read.