Non-Farm Payroll (NFP) Part I

What’s NFP?

NFP indicates the net hiring/firing of non-farmer workers for the previous month. It is published by the Bureau of Labour Statistics, usually on the first Friday of the month, in the Employment Situation Report. Some traders would lament that the attention shown on a headline figure is overstated.

NFP headline and report significance really depends on whether:

  1. Because its the earliest macroeconomic employment data to be reported, it can be considered leading. Its surprise/disappointment (levels or change don’t matter, but its the difference in actual number and forecasted number, which the market has already priced in before the release that affects the markets e.g. Trade weighted US Dollar) does have some influence on the future macro data as well… which brings us to look for contradictions between hawish/dovish NFP against other employment data like…
  2. Especially so, when monetary policymakers are looking forward to data-driven decisions on when is it timely to hike FFTR along with its narrow floor-ceiling (i.e. too early should delay (dovish)… or we need it immediately!!) and how many times is appropriate (i.e what the target rate should be) How it is being executed to keep the yields in target range depends on the innovation of the Fed Committee. But for every leveraging of money supply, there could be more than one way of deleveraging when it is time to reverse the monetary cycle. For example, in its weekly auction-buyback of treasuries, reducing its buyback is considered deleveraging its balance sheet asset (loans to needy corporate) Hmm, lending anonymously to the shadow banking and giving favored lower rates to important GDP drivers is a minority concern. Look here to get a guidance on the odds of a FFTR hike/cut in the following few FOMC meetings. Short-term rate futures are quoted as $100-3m LIBOR (now Jan, FOMC in focus is Jan + 3m)
  3. Breakdown of NFP data show give a lead to equities analyst on which sectors is organically growing and increasing their hiring. This will improve the equities sector expected EPS forecast. The list of sectors from the report include professional and business services, health care, financial activities, mining, construction, manufacturing, wholesale trade, retail trade, transportation and warehousing, information, and leisure and hospitality.
  4. Seasonality does affect the chance of your job application too.. May has the most hiring of all months. So if May’s net hiring falls and there is a negative revision to previous month net hiring figure, it is much worst than it seems!
  5. Revisions is the bane of NFP for traders. It is partly seasonal too, such as August where college students leave their summer interns and return to college
  6. In the long-term business cycle, NFP is usually underestimated at early stages of a recovery and overestimated (more disappointment or job losses) during onset of recession. Correction to this is to expect higher net hiring on coming out of recession, vice versa.
  7. Unemployment rate is from household survey, whereas NFP is from establishment survey, which is a completely different survey and hence can cause divergence between NFP change and unemployment rate. It has been the case that people trust the NFP more than unemployment data because it has a large sample size and is more reliable.
  8. Other employment statistics that matter:
  • average workweek in hours – a boss cutting cost in a recession will prefer to hire temporary workers, such as working 2 out of 5 weekdays
  • average hourly earnings – earns more, consumes more

    **Note: someone can be hired full time but paid little such that he consumes no different from unemployed or part-timers. Similarly, more could be employed but total potential contribution to the economy can be unchanged if they had compensation cuts. Hence, workers contribution to the economy is limited to his net earnings. A simple formula to estimate is:Total Unemployment Change = Δ(E x W/average weekly earnings) + H

    E = average hourly earnings
    W= workweek in hours
    H = Headline number

  • marginally attached to the labor force – highlights how bad the job search friction is.. Neither included in unemployment rate (U3) nor discouraged worker which tends to understate true employment rate (U5 includes them)

    The Headline Rate: U3

    The official unemployment rate is known as U3. It defines unemployed people as those who are willing and available to work, and who have actively sought work within the past four weeks. Those with temporary, part-time or full-time jobs are considered employed, as are those who perform at least 15 hours of unpaid family work.

  • part-time for economic reasons – involuntary part timers who prefer full time
  • participation rate – includes those employed, unemployed and those who cannot work (e.g. students, retirees). Usually its benchmark against 100% -unemployment rate to determine the contribution to economy’s consumption from students, retirees, etc.

Summary Statistics

An overall trend of the workers headcount in each sector.

n1n2n3

There is obvious seasonality in the net change in each sectors change of no. of workers.

n4

Digging deeper, we realise that the contribution to NFP headline (change) variation comes mainly from Retail Trade sector.

n5

Indeed, by grouping them into respective sectors’s change in headcounts, retail trade sector contributes the most variation, followed by wholesale trade, financial activities and manufacturing (given the shift from manufacturing to service sector in US along the decades). If you are intending to enter into one of these sector’s workforce hah, do look closer to which is the best or worst month to enter. Financial service workers life have been increasingly unstable over time 😦 exhibits heteroscedasticity)

n6

n8

n7n11n12n9n10n13n14

Forecasting (Part II)… next post

  • Assuming that human capital mobility is inflexible across industries, monthly change in each sectors net hiring should be persistent. Total variance in NFP headline can be broken down into sum of variance in each sector’s net hiring. We can thus forecast with averaging on stable component, and reduce model variation by only using models with high sample variation (e.g. GARCH) for data with strong unexpected variation where simple averaging might not be useful.
  • Handle seasonality with +/- residuals after detrending using a smoothed non-linear moving average instead of an inflexible linear trend
  • Bottom up approach: Since we know that, in ascending order variation, retail trade, wholesale trade, manufacturing, financial activities (heteroscedasticity over time) sectors’ net hiring are unstable over time, and that hiring/firing someone takes about one month, we can use previous month’s economic data to forecast the next month NFP. Namely.. [where T = current NFP reporting for previous month]
  1. ADP NFP [T-2] – as last confirmation before this month NFP
  2. JOLTS Jobs Openings – report lags by 2 months and its not timely. But is there any added value? (can be expanded to similar sectors as NFP)
  3. jobless claims [weekly] – more claims by those non-participating workers means either more unemployment or higher cost of living
  4. ISM & Philly Fed Manufacturing Index/Chicago Fed National Activity Diffusion indexes (there is for inventories, future CAPEX), State leading indexes [~ T+6] – for next month manufacturing sector’s net hiring
  5. Retail Sales & CPI [~ T+5] & Manufacturing Sales/Manufacturers: inventories to sales ratio [2m lag]- for next month retail sectors’s net hiring (can be expanded to automobile manufacturing, etc)
Advertisements

One thought on “Non-Farm Payroll (NFP) Part I

Add yours

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

A WordPress.com Website.

Up ↑

REFLEXIVE MACRO

A global macro trading blog with an emphasis on human behaviour and reflexivity.

LPL Financial Research

Macro Market Movers Blog

Market Philosopher

!# Independent Journalist, Macro Surveillance, Finance Perspectives

Curve Advisor

!# Independent Journalist, Macro Surveillance, Finance Perspectives

Joe Benning Photography

!# Independent Journalist, Macro Surveillance, Finance Perspectives

WeeklyStorybook

!# Independent Journalist, Macro Surveillance, Finance Perspectives

Fermat's Last Spreadsheet

Maths & Trading & Finance, Computing & Calculating & Coding, Languages & Learning

Reflections on Monetary Economics

!# Independent Journalist, Macro Surveillance, Finance Perspectives

The Center of the Universe

St Croix, United States Virgin Islands

The Case For Concerted Action

Post-Keynesian Ideas For A Crisis That Conventional Remedies Cannot Resolve

Tim Duy's Fed Watch

!# Independent Journalist, Macro Surveillance, Finance Perspectives

Across the Curve

A daily bond market chronicle

Pension Pulse

!# Independent Journalist, Macro Surveillance, Finance Perspectives

mainly macro

!# Independent Journalist, Macro Surveillance, Finance Perspectives

Yet Another Blog in Statistical Computing

I can calculate the motion of heavenly bodies but not the madness of people. -Isaac Newton

Macro, credit & FX Markets

!# Independent Journalist, Macro Surveillance, Finance Perspectives

Snake Hole Lounge

Economic and Financial Analysis Without The Snake Juice

Brett Dot Net

.NET & VBA

Investing For A Living

The path to worry free investing

Zero Hedge

!# Independent Journalist, Macro Surveillance, Finance Perspectives

Blog - BLACKARBS LLC

!# Independent Journalist, Macro Surveillance, Finance Perspectives

Gemma’s Bigger Bolder Baking

Baking Videos & Recipes from Chef Gemma Stafford

DanielFoodDiary.com

!# Independent Journalist, Macro Surveillance, Finance Perspectives

Quantocracy

Quant Blog Mashup

Quant at Risk

Quantitative Analysis, Risk Management, Modelling, Algo-Trading, Blockchain for Finance

The Baseline Scenario

What happened to the global economy and what we can do about it

Econbrowser

!# Independent Journalist, Macro Surveillance, Finance Perspectives

%d bloggers like this: