Events to look out for on Thursday include the US inflation reading, jobless claims data, updates on the impact of Hurricane Milton and quarterly results from Delta Air Lines:

Inflation: Price pressures in the US are forecast to have eased in September, with that month’s annual rate of inflation projected to be 2.3 per cent, from 2.5 per cent in August. Annual growth in “core” CPI, which excludes volatile energy and food prices, is expected to remain at 3.2 per cent.

Initial jobless claims: New applications for unemployment aid are expected to have ticked up to 230,000 in the week ending October 5, from 225,000 the week before.

Hurricane Milton: Officials will begin surveying the damage wreaked by the storm, which made landfall near Siesta Key in Sarasota County on Wednesday night. The National Hurricane Center earlier in the day warned that a “life-threatening storm surge, damaging winds, and flooding rains” were expected “across portions of central and south-western Florida”.

Delta Air Lines: Investors will be keen to understand the impact on the Atlanta-based airline of July’s global IT outage, which led to thousands of flight cancellations. The company is expected to post third-quarter revenues of $14.7bn before the opening bell, a less than 1 per cent increase compared with the same period last year.

How This Poll Was Conducted

Here are the key things to know about this poll:

• Interviewers spoke with 656 voters in Montana from Oct. 5 to 8.

• Times/Siena polls are conducted by telephone, using live interviewers, in both English and Spanish. Overall, about 97 percent of respondents were contacted on a cellphone for these polls.

• Voters are selected for the survey from a list of registered voters. The list contains information on the demographic characteristics of every registered voter, allowing us to make sure we reach the right number of voters of each party, race and region. For this poll, interviewers placed nearly 55,000 calls to nearly 30,000 voters.

• To further ensure that the results reflect the entire voting population, not just those willing to take a poll, we give more weight to respondents from demographic groups that are underrepresented among survey respondents, like people without a college degree. You can see more information about the characteristics of our respondents and the weighted sample at the bottom of the page, under “Composition of the Sample.”

• The margin of sampling error among likely voters is about plus or minus four percentage points. In theory, this means that the results should reflect the views of the overall population most of the time, though many other challenges create additional sources of error. When the difference between two values is computed — such as a candidate’s lead in a race — the margin of error is twice as large.

If you want to read more about how and why the Times/Siena Poll is conducted, you can see answers to frequently asked questions and submit your own questions here.

Full Methodology

The New York Times/Siena College poll of 656 voters in Montana was conducted in English on cellular and landline telephones from Oct. 5 to 8.

The margin of sampling error among the likely electorate is plus or minus 4.3 percentage points.

Sample

The survey is a response-rate-adjusted stratified sample of registered voters taken from the voter file maintained by L2, a nonpartisan voter-file vendor, and supplemented with additional voter-file-matched cellular telephone numbers from Marketing Systems Group. The sample was selected by The New York Times in multiple steps to account for differential telephone coverage, nonresponse and significant variation in the productivity of telephone numbers by state.

To adjust for noncoverage bias, the L2 voter file for each state was stratified by statehouse district, party, race, gender, marital status, household size, turnout history, age and homeownership. The proportion of registrants with a telephone number and the mean expected response rate were calculated for each stratum. The mean expected response rate was based on a model of unit nonresponse in prior Times/Siena surveys. The initial selection weight was equal to the reciprocal of a stratum’s mean telephone coverage and modeled response rate. For respondents with multiple telephone numbers on the L2 file, or with differing numbers from L2 and Marketing Systems Group, the number with the highest modeled response rate was selected.

Fielding

The sample was stratified according to political party, race and region. Marketing Systems Group screened the sample to ensure that the cellular telephone numbers were active, and the Siena College Research Institute fielded the poll, with additional fieldwork by ReconMR, the Public Opinion Research Laboratory at the University of North Florida, the Institute for Policy and Opinion Research at Roanoke College, the Center for Public Opinion and Policy Research at Winthrop University in South Carolina and the Survey Center at University of New Hampshire. Interviewers asked for the person named on the voter file and ended the interview if the intended respondent was not available. Overall, 97 percent of respondents were reached on a cellular telephone.

An interview was determined to be complete for the purposes of inclusion in the questions about whom the respondent would vote for if the respondent did not drop out of the survey after being asked the two self-reported variables used in weighting — age and education — and answered at least one of the questions about age, education or presidential-election candidate preference.

Weighting (registered voters)

The survey was weighted by The Times using the survey package in R in multiple steps.

First, the sample was adjusted for unequal probability of selection by stratum.

Second, each poll was weighted to match voter file-based parameters for the characteristics of registered voters.

The following targets were used:

• Six categories of partisanship (Classification based on an NYT model of vote choice in prior Times/Siena polls)

• Partisanship (L2 model based on commercial data and partisan political contributions)

• Race or ethnicity (L2 model)

• Age (self-reported age, or voter-file age if the respondent refused) by gender (L2 data)

• Education (four categories of self-reported education level, weighted to match NYT-based targets derived from Times/Siena polls, census data and the L2 voter file)

• White/nonwhite race by college or noncollege educational attainment (L2 model of race weighted to match NYT-based targets for self-reported education), if part of the non-Black-or-Hispanic sample

• Marital status (L2 model)

• Homeownership (L2 model)

• Turnout history (NYT classifications based on L2 data)

• Method of voting in the 2020 elections (NYT classifications based on L2 data)

• State region (NYT classifications)

• Census block group density (A.C.S. 5-Year Census Block Group data)

Finally, the sample of respondents who completed all questions in the survey was weighted identically as well as to the result for the general-election horse-race question (including voters leaning a certain way) on the full sample.

Weighting (likely electorate)

The survey was weighted by The Times using the R survey package in multiple steps.

First, the samples were adjusted for unequal probability of selection by stratum.

Second, the first-stage weight was adjusted to account for the probability that a registrant would vote in the 2024 election, based on a model of turnout in the 2020 election.

Third, the sample was weighted to match targets for the composition of the likely electorate. The targets for the composition of the likely electorate were derived by aggregating the individual-level turnout estimates described in the previous step for registrants on the L2 voter file. The categories used in weighting were the same as those previously mentioned for registered voters.

Fourth, the initial likely electorate weight was adjusted to incorporate self-reported intention to vote. Four-fifths of the final probability that a registrant would vote in the 2024 election was based on the registrant’s ex ante modeled turnout score, and one-fifth was based on self-reported intentions, based on prior Times/Siena polls, including a penalty to account for the tendency of survey respondents to turn out at higher rates than nonrespondents. The final likely electorate weight was equal to the modeled electorate rake weight, multiplied by the final turnout probability and divided by the ex ante modeled turnout probability.

Finally, the sample of respondents who completed all questions in the survey was weighted identically as well as to the result for the general election horse-race question (including leaners) on the full sample.

The margin of error accounts for the survey’s design effect, a measure of the loss of statistical power due to survey design and weighting.

The design effect for the full sample is 1.24 for the likely electorate in Montana.

Among registered voters, the margin of sampling error is plus or minus 4.3 points in Montana, including a design effect of 1.26.

For the sample of completed interviews, among the likely electorate, the margin of sampling error is plus or minus 4.5 points in Montana, including a design effect of 1.29.

Historically, The Times/Siena Poll’s error at the 95th percentile has been plus or minus 5.1 percentage points in surveys taken over the final three weeks before an election. Real-world error includes sources of error beyond sampling error, such as nonresponse bias, coverage error, late shifts among undecided voters and error in estimating the composition of the electorate.

Unlock the Editor’s Digest for free

Roula Khalaf, Editor of the FT, selects her favourite stories in this weekly newsletter.

At the end of September there was a big spike in the Secured Overnight Financing Rate. This may already be putting you to sleep but it’s potentially a big deal, so please stick around.

SOFR was created to replace Libor (R.I.P.). It measures the cost of borrowing cash overnight, collateralised with US Treasuries, using actual transactions as opposed to Libor’s more manipulation-prone vibes. You can think of it as a proxy of how tight money is at any given time.

Here you can see how SOFR generally traded around the central point of the Federal Reserve’s interest rate corridor, and fell when the Fed cut rates by 50 basis points in September. But on the last day of the month, it suddenly spiked.

This is natural, to an extent. There’s often a bit of money tightness around the end of the quarters, and especially the end of the year, as banks are keen to look as lean as possible heading into reporting dates. So SOFR (and other measures of funding costs) will often spike a little around then.

But this was FAR bigger than normal. Here is the same chart but showing the end-of-2023 spike, and little dimples at the end of the first and second quarters.

Indeed, Bank of America’s Mark Cabana estimates that this was the single-biggest SOFR spike since Covid-19 wracked markets in early 2020, and points out it happened on record trading volumes.

Cabana says he was initially too hasty in dismissing the spike as driven by a short-term collateral shortage and unusually large amounts of window-dressing by banks. In a note published yesterday, he admits to overlooking something potentially more ominous: reserves seeping out of the banking system.

We have long believed funding markets are determined by 3 key fundamentals: cash, collateral, & dealer sheet capacity. We attributed last week’s funding spike to the latter 2 factors. We overlooked extent of cash drain in contributing to the pressure.

The increased sensitivity of cash to SOFR hints of LCLOR.

LCLOR stands for “lowest comfortable level of reserves”, and might require a bit more explanation.

Back in ye olde times (pre 2008), the Fed set rates by managing the amount of reserves sloshing around the US monetary system. But since 2008 that has been impossible due to the amount of money pumped in through various quantitative easing programmes. That has forced the Fed to use new tools — like interest on overnight reserves — to manage rates in what economists call the “abundant reserve regime”.

But the Fed has now been engaging in reverse-QE — or “quantitative tightening” — by shrinking its balance sheet sharply since 2022.

The goal is not to get the balance sheet back to pre-2008 levels. The US economy and financial system is far larger than it was then, and the new monetary tools have worked well.

The Fed just wants to get from an “abundant” reserve regime to an “ample” or “comfortable” one. The problem is that no one really knows exactly when that happens.

As Cabana writes (with FT Alphaville’s emphasis in bold below):

Like the macro neutral rate, LCLOR is only observed near to or after it is reached. We have long believed LCLOR is around $3-3.25tn given (1) bank willingness to compete for large time deposits (2) reserve / GDP metrics. Recent funding vol supports this.

A similar dynamic was seen in ‘19. At that time, the correlation of changes in reserves to SOFR-IORB turned similarly negative. The sensitivity of SOFR to reserves correlation signalled nearing LCLOR. We sense a similar dynamic is present today.

Unfortunately, when reserve levels drop to uncomfortable levels, we tend to find out very quickly, in unpleasant ways.

Cabana’s mention of 2019 is a reference to a repo market crisis in September that year, when the Fed missed growing hints of tightness in money markets. Eventually it forced the Federal Reserve to inject billions of dollars back into the system to prevent a broader calamity. MainFT wrote a superb explainer of the event, which you can read here.

In other words, the recent SOFR spike could be a hint that we are approaching or already in uncomfortable reserve levels, which could cause a repeat of the September 2019 repo ructions if the Fed doesn’t act preemptively to soothe stresses.

Here are Cabana’s conclusions (his emphasis):

Repo is heart of markets. EKG measures heart rate & rhythm. Repo EKG flags shift. Cash drain has supported spike in repo. Fed should take repo pulse & sense shift. If Fed too late to diagnose, ‘19 repeat. Bottom line: stay short spreads w/Fed behind on diagnosis.