To The Moon

Core CPI: 2.6%. Core PCE, the Fed's own gauge: 3.0%. Core CPI over the last three months, annualised: 3.4%.

Three "core" numbers, one rate decision. The gap between them is the story. It is not the headline number minus the funds rate.

The Fed decides today (usually 2 PM ET, which is 1 AM Thursday in Bangkok, 2 AM in Singapore). I am not predicting the decision. I want to show how I read the debate. The method works on any inflation scare.

1) Which "real rate"?

The funds midpoint is 3.875%. Subtract headline CPI (3.7%) and you get 0.2. Subtract core CPI (2.6%) and you get 1.3. That is just arithmetic. It hides a choice.

The choice is the gauge. The Fed targets PCE, not CPI. Normally core PCE runs below core CPI. From 1990 to 2019 it did in 86% of months, by a median of 0.47 points (FRED data). In 2026 it has run above core CPI in every month so far, by 0.3 to 0.6 points. That is a flip.

The September minutes give a clue. A few participants said software and portfolio management fees had made large contributions to PCE readings. Portfolio fees are charged as a share of assets, so they rise with stock prices. My hypothesis: a rising market has been feeding the Fed's own inflation gauge. I could be wrong. Shelter is 36% of CPI, and I have not checked its smaller weight in PCE. That may explain part of the gap.

One trap: the BEA changed its method on 30 September. The minutes quote core PCE at 3.4%. The revised August figure is 3.0%.

2) Read the run-rate, not the year

Core CPI is 2.57% over 12 months. Its last three monthly prints (0.2, 0.3 and 0.3) annualise to 3.4%. The 12-month number is slow to turn. The 3-month number is quicker, but noisy; the minutes say so.

A reusable test: ask what the next print must be to pull the 3-month rate back to the 12-month level. For October core CPI that is about 0.0%. A 0.2% print keeps it near 3.4%.

3) The strongest case against me, and a history lesson

Past energy shocks were usually looked through. I checked 10 episodes since 1960 where headline CPI ran at least 1 point above core and the gap was widening. After 1982, in 7 of 7, the gap closed because headline fell to core. Before 1982, in 3 of 3, core rose to meet headline.

So the real question is which regime this is. Look-through was tested when underlying inflation was low: the March 2011 statement called energy effects transitory while measures of underlying inflation were "somewhat low". Now PCE inflation has been above 2% for 66 straight months, and the price level sits about 11% above a 2% path (FRED). The Fed did not look through this time. The September vote was 12-0 for a hike, and the minutes record: "Several participants stated that they viewed the current policy rate as not restrictive or only mildly restrictive."

Limits: ten cases, my own cut-offs, 2008 was a demand collapse, and 2022 followed 525 bp of hikes. It is a way to think, not proof.

Two more things. First, the Fed's September projections hold the rate at 4.1% from end-2026 to end-2027 while PCE inflation falls from 3.7% to 2.3%. The whole rise in the real rate comes from inflation falling, none from rate moves. Second, energy jumped 10.9% in March 2026 (BLS, seasonally adjusted). That month drops out of the 12-month figure in March 2027. If energy simply stayed flat, I get energy's contribution to headline CPI falling by about 0.75 points in that single print. My hypothesis: a headline drop next spring will not prove oil fell or that policy worked.

4) What this does to a put seller

Put-call parity says the call price minus the put price equals the interest on the strike. At 4% for 30 days that is 0.33 per 100 of strike. So higher cash yields do not make puts richer. They move part of your pay from premium to collateral interest.

Hypothetical numbers, per 100 of strike, 30 days, cash at 4%:

- 5% out-of-the-money put at 20% IV: premium 0.51, interest on the collateral 0.31.

- 10% out-of-the-money put at 25% IV: premium 0.19, interest on the collateral 0.30.

If the collateral sits idle and earns nothing, that interest is what you give up. Further out of the money, interest dominates the premium. Check what your broker pays on idle cash. Tiger Vault advertised "up to 3.7%" as of 7 April 2026; I have not checked today's figure, so look at the live 7-day yield in the app.

Break-even realised volatility for an at-the-money put sold at 25% IV is 25.0%, whether cash pays 0% or 4%. The rate does not move it. What moves it is how far stocks drift above cash: 5 points above cash lifts it to 26.7%.

(Black-Scholes, flat vol, no dividends. Real skew differs.)

5) Singapore

MAS steers the exchange-rate band, not a policy rate. On 14 October it left the slope unchanged, after raising it in April and July. September CPI is 2.2%, MAS core 2.0%. The 6-month T-bill cut-off was 1.95% on 22 October. US 3-month yield is 4.22%.

So SGD cash is about 0.25 below Singapore headline inflation. USD cash is about 0.5 above US headline. The difference is paid for in the exchange rate.

Made-up units: 1,000 units of collateral for 30 days. USD at 4% earns 3.3. SGD at 2% earns 1.6. The gap is 0.16%. A 1% rise in SGD against USD costs the USD holder 10 units. Over three years, USD/SGD moved by more than 0.17% in 90% of 21-day windows (FRED). The exchange-rate noise is bigger than the yield gap. (USD/SGD is not the S$NEER, a trade-weighted basket.)

One more Singapore link. Electricity tariffs follow gas prices with a one-quarter lag. SP Group cut the Q4 household tariff 10.4%, and said a rise could follow if fuel prices stay up. So Singapore's headline lags the energy move.

Where is the energy shock priced? In oil volatility, not index volatility. OVX is 58.7 (92nd percentile since 2007). The VIX is 18.4 (55th). The ratio is 3.3 this year against about 2.1 in 2019 to 2025. I do not claim oil vol feeds equity vol; the day-to-day correlation is unchanged.

What I am not claiming: a view on tonight's decision or on oil, proof from ten episodes, or a rule for which strike to sell.

What would change my mind: October core CPI at 0.1% or less (10 Nov, 8:30 AM ET = 9:30 PM Singapore, 8:30 PM Bangkok), or September core PCE (29 Oct, 8:30 AM ET = 7:30 PM Bangkok) at or below 3.0%.

Later: December's projections, the March 2027 CPI (energy base effect), and Singapore's Q1 2027 electricity tariff around end-December (date unchecked).

Which of the three "core" numbers do you read first, and why?

Educational, not advice. I may hold positions in the names mentioned.

Stop guessing. Start calculating.

Live to fight another day. 🤙

#options #tradingpsychology #riskmanagement #optionstrading