Soybean Basis and Spreads Implications for Marketing ...€¦ · Typical: Stock/use vs price...

transcript

Soybean Basis and Spreads

Implications for Marketing Decisions

To ND Soy Council Marketing Conference, Fargo

William W. Wilson

Dept. of Agribusiness and Applied Economics

Basis Behavior –

◦ Concepts, Seasonality and Behavior

◦ Intermonth spreads

◦ Impacts of shipping costs

◦ Forecasting Basis

◦ Market year analogues for export basis

◦ Seasonal analogues

Typical: Stock/use vs price

2018/19: suggest prices 700-800c/b and take 4-6 years to return to normal

2019/20: Revised to .11 and now .104 prices about 875c

2020/21 Revised to .07 implies higher futures approaching 10$

Dept of Agribusiness & Applied

Economics,NDSU, Fargo - 58102 3

Taxonomy of Markets

Futures

B = C - F

Forward

-specific terms of

trade (location, time,

quantity)

Forward

-standardized terms

of trade

Arbitrage

Management

Rail Markets

Forward

Review of Basis

High Correlation Market B=C-F

Key Points:

1) correlation

2) risk: St deviation

Correlation =

Basis Standard Deviation =

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

Period

Cash Futures

$(0.35)

$(0.30)

$(0.25)

$(0.20)

$(0.15)

$(0.10)

$(0.05)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

PeriodP

Review of Basis

Low Correlation Market

Correlation =

Basis Standard Deviation =

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

Period

Cash Futures

$(0.30)

$(0.20)

$(0.10)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

PeriodP

Factors impacting basis

Location: Shipping

Time: Storage cost

Quality: Premium/discounts

Delivery

D = Storage

Harvest

Overall Behavior of basis:

Reduced Cost of Storage and Delivery Month

Behavior

Recent basis behavior

Jamestown

PNW Track

Brazil

MyDTN Basis for soybean 4 14

Jamestown Seasonal Soybean Basis Current: new crop -120 (65c/b under normal)

Kensal area Basis:Normal=-65 Now -110

Arthur (Ayr): 2019/20 about

50C/b<normal

Ayr: Extreme post-

harvest appreciation

Seasonal PNW Soybean BasisCurrent: Spot +80 (or less) vs. (vs average 130)

Most important factors impacting US Export Basis:◦ Brazil Export Basis

◦ China imports

◦ Etc.

Recent study: Bullock and Wilson, Factors impacting the US Export Soybean Basis (forthcoming)

PNW Basis update

Brazil Soybean Basis

Impacts of China Tariffs:

25% is about 190c/b

Impact◦ Delivered price China for US is increased 190c/b relative to all

competitors

◦ Value of Brazil, and all other potential exporters, increases toward this value, 190c/b

◦ Taken together, the world has a bifurcated market Those markets (China) with tariff on US imports

Those markets (all others) without tariff

Compounded by China Gov. Purchases◦ COF CO and Sinograin: Purchases made by these organizations,

and even from US are effectively tariff-free

◦ Commercial buyers in China do not have this advantage

◦ Impact: Purchase made through COFCO and Sinograin to provide good-faith to trade negotiations.

◦ Some may be cancelled, and/or switched to further months forward

Bzl would be at a $70 (190c/b) discount to the US.

bzl basis peaks in September/October and decreases rapidly there after.

Soybean Basis: Gulf and PNW

Basis values at U.S. Gulf and PNW are highly correlated

PNW is normally a premium to US Gulf reflecting the greater shipping costs to that port; and the lower ocean shipping costs from that port

Current: FOB Basis

Spatial Competition for Northern

Soybean Jan 15 2020, and 4 9 2020

Intermonth spreads

Price of Storage Discuss

◦ Inverted and normal markets

◦ Theory of the price of storage

Illustrate

◦ Normal and inverted

◦ Futures, basis, logistics: Inverse/normal could occur in any of these markets\

POINT: Intermonth spread and total spread represent the ‘market determined price of

storage’ Ps, vs storage cost S

◦ If Ps >S store (store and sell futures e.g., storage hedge)

◦ Ps <S store (don’t store; sell cash)

Futures

Normal Market Inverted Market

Intermonth

Spread

Intermonth

Spread

March 800 800

May 820 +20 775 -25

July 835 +15 +35 745 -30 -65

Supply and Price of Storage

Elements of the S or Storage:

◦ Downward sloping convenience yield (ct)

◦ Upward sloping risk premium (rt)

Large crops/stocks results in

◦ large intermonth storage costs; or,

◦ discounts for nearby relative to deferred delivery periods

Congestion◦ As shipping demand (for all/any commodities)

increases, congestion accrues, resulting in increases in shipping costs

Congestion costs can:

◦ Increase to very large levels

◦ Increase sufficient to divert shipments into storage, or to other routes/modes

◦ Common in Ports

Port congestion

Rail capacity

Crucial in international competition

Convenience Yield

Physical Cost

Risk Premium

Supply

Stocks

Price Of Storage

3.Supply and Demand for Storage

Downward sloping demand

◦ Inventory (K) will result in an

inverted market (Q)

◦ Inventory (L) will result in a

carry market (R)

Current Soybean Market:

April 13, 2020

Simple:

◦ Narrower carry

◦ Store

◦ Sell nearby

Soybean Prices: Apr 2020

2020Futures Diff Cumu

lative

Months Per

Return to

Storage

Cost to Store 5

May 854

July 862 8 8 2 4 -2

August 865 3 11 3 4 -4

Sept 866 0.75 11.75 4 3 -8.25

Nov 870 4.5 16.25 6 3 -13.75

Jan-20 873 3 19.25 8 2 -20.75

FUTURES SPREADS ONLY! Basis declines Jan-July

Price of Storage

Forward Curve

Definition:

◦ Curve representing the prices at which

the market is willing to transact future

business today.

◦ Reflects normal vs inverted markets

2020 Soybean:

Forward Curves CME Soybeans Normal Market

DTN: Cost of storage calculator

Logistics

Rail Rates and Car Markets

Elements of rail shipping costs

◦ Tariff Rates (below)

◦ Car markets: Primary and secondary

(Below)

◦ FSC---now cancelled

Rail Car Markets

Primary car auctions are at nil premium (avg $54/car)

Secondary car values (DCV)

◦ More risky

◦ Historical: 4 peaks 2006, 2008, 2010, 2013

◦ Sharp increase in 2013/14, and, sharp decrease in recent weeks

◦ Large premiums allocate demand and storage (i.e. don’t ship!)

◦ Now: converge to 0 or <0 premium!

Shippers

◦ Have to option to take coverage

◦ Not all shippers are covered; some benefit while others are out of the market

Railroad Pricing Mechanisms

Rail Car Market: Taxonomy

Primary market: Initial rail car auction from RR

◦ Weekly

◦ 12-24 months forward

◦ Other terms

Secondary market:

◦ Brokerage

◦ Offers from buyers (e.g., CHS, etc.) as part of procurement

GrainCO: has 3 ways to make trades of trains in secondary transactions◦ sold on a trip-by-trip basis to other

companies;

◦ used by company owned elevators; and

◦ forward contracted to other companies for a set quantity, delivery period and duration.

GrainCO: in trades to affiliated and/or target supplier

◦ typically sells their shuttles to elevators for $25-$50/car/trip over the premium that they paid the carrier.

◦ assumes all risk and liability in regard to the cars being placed on time.

BNSF Primary Auction for COTs

BNSF Primary Auction for COTs 11

2nd Car Market

Tradewest 2nd Market August 2017--typical

2nd Market Feb 11 2019

Nearby values

inflated

Likely due to

◦ Service problems

◦ Anticipated/real

nearby exports

soybean and HRS

Wide B/A spread

Inverted

New crop nil offers

Recent 2nd market Values 4 13 2020

USDA Grain Transportation

http://www.ams.usda.gov/AMSv1.0/ge

tfile?dDocName=STELPRDC5069058

BIG DEAL: Improve transparency

USDA Oct 2018

Company offers in 2nd transactions

Data Behavior

For illustration

As used, in most

cases below in

empirical models

Expected Quantity per Shuttle—

Primary Market Shipments per month defined as:

◦ V=Velocity per month

E.g., 2.5 trips per month

V is random

◦ Qshuttle= # shuttles bought in

primary market

E.g. 10*V=25 trains (110 cars)

per month

◦ Cars: 100-110 rail option

Role of 2nd market

◦ If V*Q>shipping demand: sell on

2nd market

◦ If V*Q<shipping demand: buy on

2nd market

Velocity (See data below)

Actual cars derived as:

Ca=V*To*110

Ca = actual cars

V=Velocity

To= Trains ordered

110= cars/train (actual = [100,110] rail option

Conversion: 3750b/c

Velocity

Trips/month 1 2 3 1 2 3

2.7 2.7 5.4 8.1 1,113,750 2,227,500 3,341,250

2.8 2.8 5.6 8.4 1,155,000 2,310,000 3,465,000

2.9 2.9 5.8 8.7 1,196,250 2,392,500 3,588,750

3 3 6 9 1,237,500 2,475,000 3,712,500

3.1 3.1 6.2 9.3 1,278,750 2,557,500 3,836,250

3.2 3.2 6.4 9.6 1,320,000 2,640,000 3,960,000

3.3 3.3 6.6 9.9 1,361,250 2,722,500 4,083,750

Trains bought

Car/month

Trains bought

Bushels/month

Velocity

Correlations—Very Interesting!

Correlation Basis PNW

Jamestown ND

Jamestown

Spread

Nearby Soybean

Futures DCV ($/car)

Basis PNW 1.000

Jamestown ND 0.311 1.000

Jamestown

Spread 0.777 -0.219 1.000

Nearby

Soybean

Futures 0.514 0.090 0.490 1.000

DCV ($/car) 0.500 -0.199 0.589 0.251 1.000

Rail Car Market: Summary

Feature Primary Secondary

Allocation

mechanism

Bid to buy from Railroad Acquire from brokers or

other shippers

Quantity

allocated

Bid on 12, 24, 36 months of

trains by month

Offers by other shippers

Tradable Yes Yes

Quantity Receive X*N where X is rail

cycle (e.g., 2.9 trips/month

random) and N is number of

trains

Amount purchased

Car placement By period (10 day) s.t. car cycles Assured, by period (10

Guarantee None Yes, by month

Cancellation Possible s.t. penalty None (re-sell in 2nd

market)

Price auction: Average=54$/c s.t. min Market: Average=$225;

no min/max

Grain Pricing I:

Basis derivation Implications

Define: B=basis, c=cash, f=futures, T=tariff rail rate;M=margin; d1,d2 are 2 different destiatins, FSC is fuel service

charge; Car is car premium

1 Bo=Co-FConventional: Basis is constant; highly

predictable;

2 Bo=Bd-M-T-F Little more complex

3 Bo=MAX[(Bd1-T1), [(Bd2-T2), [(Bd3-

T3)] -M-F

Multi-markets: makes basis at origin more

dependent on basis and shipping costs to

multiple markets. More volatility in basis!

4 Bo=Bd-[Rt-FSC-Car]-M-F Numerous rail mechanisms makes greater

uncertainty in some elements of shipping costs;

for those shippers not covered, values have

greater volatility (risk)

5 Bo,t+n=Bd,t+n -

[Rt+n-FSCt+n-Cart+n]*-(M+RP)-Ft+n

Forward transaction results in volatility in basis

values resulting from un-covered shippers

having to infer expectations of relevant values;

including an implicit risk premium in margins

Summary

US Grain Shipping has evolved radically in past 2 decades

◦ Greater flexibility

◦ Greater efficiency in shipping

◦ Declining pressure on shipping rates

3 components to rail pricing

◦ Tariff

◦ FSC (if applied)

◦ Shuttle premium

Implications for traders

◦ Knowledge of the market

◦ Integrate commodity and freight positions: Long commodity/short freight Evaluate risk of freight positions

◦ Forward purchases reduce risks for suppliers and facilitates better offers

◦ A portion of the seasonal price (futures and basis) is partially offset by shuttle premiums and barge rates

Break: QUestions

Basis Analytics

SOurces

GeoGrain

DTN-ProphetX

Mapping

Projections

◦ Seasonal

◦ MACD

Spatial competition

Grain Hedge: Basis Soybeans

MyDTN Basis for soybean 4 14

Projections: Short term and

Seasonal

Shorter term:

◦ Simple comparisons of spot vs.. moving

averages

◦ Turning points! MACD

◦ Projections based on MACD (shorter term)

Seasonal indexes and current period

basis values

Seasonal Projections

Price Behavior Using Moving Averages

Kensal area Basis:Normal=-65 Now -110

Ayr: Extreme post-

harvest appreciation

Seasonal Analysis and Simple

Projections

• Seasonal Motivations

• Concept of seasonally produced crops

• Seasonal factors affecting basis behavior

• Seasonal Indexes

• Derivation of Seasonal Indexes2

2 A moving average seasonal index was calculated using monthly data from 1985-1996. The seasonal index value

is calculated by estimating the center average of 12 months of prices, then the current price is divided by this

average price. Index scores are then averaged across years for each month to get the average seasonal index

score for a month.

Jamestown Average Soybean

Basis Seasonal basis behavior

shows that on average◦ Basis is at a bottom in July-

August

◦ Basis is at a peak in December

Seasonal changes◦ May-August: -20c

◦ August-Dec: +40c

Basis variability (risk) is shown by the st. deviations◦ Basis is least risk at and

after harvest time

◦ Basis risk escalate in months following January, no doubt due to uncertainties of the impact of Brazilian crop and basis competition

Seasonal Jamestown Soybean Basis Index

Average of 2010/11-2018/19 (Sept-August)

-1 Std. Dev.

+1 Std. Dev.

-1 Std. Dev. 0.77 0.67 0.93 1.09 1.06 1.01 0.76 0.54 0.83 0.86 0.35 0.27

Mean 0.98 1.05 1.15 1.20 1.16 1.11 1.01 0.89 0.98 0.98 0.67 0.61

+1 Std. Dev. 1.18 1.42 1.37 1.30 1.26 1.22 1.25 1.23 1.14 1.09 0.98 0.95

Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug

Seasonal Analog Derivation Using

Cluster Analysis - PNW

Factors Influencing PNW and Gulf

Export Basis

The marketing year average basis level for both Gulf and PNW is highly correlated and is

primarily influenced by competitive pressures – both international and domestic.

Most important impacting basis level:

◦ Brazil Basis

◦ Chinese imports

No single dominant seasonal pattern for either Gulf or PNW basis – the seasonality

characterization varies substantially from year-to-year and is not correlated between the two

markets.

Seasonality of the basis is primarily influenced by

◦ the level of export activity

◦ internal logistical factors such as railcar bottlenecks / delays,

◦ pace of farmer deliveries into the market,

◦ and transportation costs – primarily barge and ocean freight.

Break: QUestions

Spatial Competition for Northern

Soybean

Stl: Level; and seasonal

appreciation: -40 to +30

PNW Track: Level and seasonal

appr: 25 to +90

Current Soybean: StL vs PNW Spot and New

Net Returns 4 14

Competition:

◦ Stl vs PNW

◦ Current result

Spot: PNW Dominates

◦ Basis at PNW is stronger

◦ Net returns favors shipping to PNW

New Crop

◦ Basis in both markets extremely depressed

◦ Net return slightly favor PNW, but only by 6/b (other factors could influence shipping decisions (e.g. barge, congestion, velocity)

St. Louis

Transfer

PNW Track StLouis

transfer

PNW Track

Central ND

Track Values 30 115 12 90

Freight to Track: Excl car premiums 100 153 100 153

Net Jamestown -70 -38 -88 -63

Jamestown bid -95 -95 -125 -125

Gross Return 25 57 37 62

New Crop 2020Spot 2019

Forward basis curve 11 30

Critical: Basis FOB for months forward

Competitive during nearby

Brazil basis falls sharply (70c/b) into March 2020

Like pressure will be on US values, port and interior!

Comparable FOB Basis: Current 4 13

Export fob basis 4 13 2020

International Arbitrage 4 13

Current market impacted by

◦ Cheap PNW and Bzl Basis,

◦ Cheap PNW ocean shipping vs USG

C&F Values (China) favor

Bzl gains advantage due to cheaper basis in April-August

[Exclude perceived/real quality differentials vs Bzl]– 10-20c/b

International USG PNW Bzl

FOB Port (M&P) c/b 75 115 82

Ocean shipping (US ports to China)$/mt 36 17 23

Total Basis + Ocean (delivered China) 172 160 145

Total Cost: Futures (893)+ Basis + Ocean $/mt 391 387 381

USG PNW Bzl Bzl Adv

April 374 369 364 6

May 380 375 370 6

June 380 375 370 6

July 381 377 373 3

Aug 381 377 374 2

Sept 382 377 378 -1

Oct 382 377 380 -3

Nov 385 380 383 -3

Derivation delivered China 4 13 2020

Delivered China $/mt

2020: Chinese Purchases and Coverage

Potential slots:

◦ June-Aug;

◦ New crop (US)

Gut-month for

competition:

◦ Bzl vs US

◦ Bzl seasonal low

basis to May!

◦ Bzl origins

currently favored

for these positions

China Imports: Mar 31 2020

Purchase

To date

Yet to

Covera

Low High

Sep-19 7.3 7.3 7.3 0 100

Oct 7.3 7.3 7.3 0 100

Nov 9.6 9.6 9.6 0 100

Dec 5 5 5 0 100

Jan-20 3.5 3.5 3.5 0 100

Feb 6.7 6.7 6.7 0 100

Mar 7.5 7.5 7.5 0 100

Apr 8.5 8.5 8.5 0 100

May 8.5 9 8.5 0.25 97

June 9.5 9.5 6.7 2.8 71

July 7.25 7.25 4.2 3.05 58

Aug 7 7 1.2 5.8 17

2019/20 Total 87.65 88 76 12 86

Feb-21 Estimate 7 1.50 5 22

21-Mar 8 2.00 6 27

21-Apr 9 0.80 8 9

21-May 9 0.60 8 7

21-Jun 10 0.50 9 5

21-Jul 7 0.40 7 6

2020/21 Est total 48 5.80 43 12

Demand Range

Implications for Growers

Basis provides direction for decisions related to:

Making effective hedging decisions

Basis impacts

Timing or shipping (not purchasing) decisions

Storage decisions

Quality decisions

Shipping strategies (more later)

Understanding basis is critical in negotiations with suppliers/handlers for

both flat price, basis contracts (discussed later) and premiums/discounts for

quality

Major results

Basis had been increasing, now declining in to new crop

Outlook is for lows in October, increasing into Dec/Jan and then declining

Intermonth spreads: All are inverted.

Sell and don’t store

Don’t‘ store to new crop

Sell new crop for as nearby as possible

Basis Fundamentals: 2019

Old crop: Issues China trade-deal related

purchases

US Logistics◦ USGulf flooding

◦ Rail velocity

Bzl harvest, and lower basis

Bzl logistical problems: potential shifting some sales to PNW

Normal seasonal declines Feb-July

New Crop China trade resumption

◦ Elimination of import tariffs

◦ Govt vs private buying

Bzl logistics in Aug-Oct period

Spreads in US futures

PNW dominance; vs US Gulf

US is competitive Sept forward until Bzl new crop

Intense competition in Aug-Sept◦ US marketing 2 crops

◦ Weakening of basis: -20 to 50cb to August

◦ Narrowing intermonth spreads discourage storage post harvest. Only Basis

To Revert to historically High

Values

Requirements

◦ Increase in Brazil basis

Increased shipping costs

◦ Substantial increase in China purchases

◦ Rail rate reductions (substantial)

Grain Sellers: Fundamental Market Strategies (Wilson

Interpretation)

Expectation Likelihood Best Cons. Worst

Basis/Spread Futures

High Sell Futures Buy Puts (ITM) Long cash

Low Buy Puts (OTM) Sell Futures Long Cash

High Long Cash Buy Puts

Low Long Cash Out-of-Money

Sell Forward

Cash Contracts

Buy puts

Basis contract

NPE or DPC

HTA or sell

futures

Buy puts

As confidence about the underlying increase in futures

*Increases: Shift to more direct futures of ITM Options (higher delta)

*Lessens: Shift to more options and lower OTM options (lower delta)

Soybean Basis and Spreads Implications for Marketing ...€¦ · Typical: Stock/use vs price...

Documents