← HOOD forward model

HOOD · Forward model · Net interest · Bear case

What has to happen in Net interest

Model as of

This page changes Net interest inside the complete HOOD model. Every other vertical, corporate programme and valuation assumption stays in the selected portfolio case.

HOOD forward model
Horizon
Consolidated fair value $31.63 all other verticals held in this portfolio case
Final-quarter revenue $421M 23% of company revenue
Explicit segment contribution $4.67B EBITDA less segment capex, before corporate items

Payment for order flow is capped or banned and takes the rate out of options and equities together. Sports event contracts lose their fight with state gaming regulators and the second-largest transaction line stops rather than slows. Rates fall faster than balances grow, so net interest shrinks. Crypto keeps bleeding at July's pace rather than the model's, and the multiple compresses to what a cyclical broker is worth.

Net interest

Basis quarter$389M
Final quarter$421M
Implied CAGR+2%
Final revenue mix23%

The half of the business nobody talks about: $389M a quarter earned on customer balances - a $21.6B margin book, $29.7B of cash sweep and $18.7B of customer cash. It grows with balances and shrinks with rates, and it is the reason a bad quarter for trading is no longer a bad quarter for Robinhood.

Last four quarters
2025 Q3 $456M Reported
2025 Q4 $411M Reported
2026 Q1 $359M Reported
2026 Q2 $389M Reported
Margin lendingCash sweep to partner banksSecurities lendingCorporate cash and segregated balances
Capacity energised 70 $B of balances at the basis quarter $70B of interest-earning balances: $21.6B margin book, $29.7B cash sweep, $18.7B customer cash.
Capacity added 4 $B of balances/qtr changing 0.0% per quarter $3.5B of balances a quarter, against $21.7B of net deposits in the June quarter and $5.6B in July.
Utilisation 100% gliding toward 100% 100%: every dollar in the base already earns. Move it only if you mean balances stop paying.
Revenue per $B of balances $5.56M/qtr drifting −1.0% per quarter $5.56M per $1B a quarter - $389M over $70B, about a 2.2% annualised net spread.
Net interest

Latest: $421M (2031Q2E)

Period Value
2025Q1 $290M
2025Q2 $357M
2025Q3 $456M
2025Q4 $411M
2026Q1 $359M
2026Q2 $389M
2026Q3E $394M
2026Q4E $399M
2027Q1E $404M
2027Q2E $407M
2027Q3E $411M
2027Q4E $414M
2028Q1E $417M
2028Q2E $419M
2028Q3E $421M
2028Q4E $422M
2029Q1E $424M
2029Q2E $425M
2029Q3E $425M
2029Q4E $425M
2030Q1E $425M
2030Q2E $425M
2030Q3E $424M
2030Q4E $424M
2031Q1E $422M
2031Q2E $421M

Assumptions & reasoning

  • Balances x net spread, in billions, because that is what net interest revenue physically is. A growth rate here would hide the fact that balances and rates move independently, and usually in opposite directions.
  • The $70B base is the margin book at $21.6B plus cash sweep at $29.7B plus customer cash and deposits at $18.7B, each of them a disclosed number. $389M over $70B is $5.56M per billion a quarter, about a 2.2% annualised net spread.
  • July's monthly data says the balances are flat to down, not growing: margin book $20.7B, down 4% on June, cash sweep $29.2B, down 2%. The $3.5B a quarter of balance growth assumed here is a bet on net deposits, which ran $5.6B in July alone.
  • The spread falls 1% a quarter, which is the rate-cut assumption doing its work. This is the input to change first if you disagree about the path of short rates - it is worth more to this line than balance growth is.
  • Utilisation is held at 100% because every dollar in the base is already earning something. It is a slider that should not be moved unless you mean that some of these balances stop paying.
HOOD model map

Explore another vertical