1

C A L I F O R N I A R E S E A R C H B U R E A U

Estimating Tax RevenueLosses due to Remote

SalesFederation of Tax Administrators

Revenue Estimation and TaxResearch Conference

Oklahoma City, Oklahoma

October 11, 2005

Martha Jones, Ph.D.

Outline

• Trends in e-commerce (E-stats)• Bruce/Fox and DMA sales and use tax

losses from e-commerce• Estimates by states of their own losses• Compare e-commerce loss estimates to

use tax collections– Actual use tax data– Estimated use tax data

2

Use tax

• States with a sales tax have acompensating use tax.

• Use tax definitions vary across states.• Who collects use tax? Quill, 1992

Census E-Stats Data Coverage

• 1999 - present• E-Stats data are national.• E-Stats do not cover the entire U.S. economy.• E-commerce measures Internet, EDI

(Electronic data interchange), etc.

3

E-stats are based on data collected in 5 surveys(not additive)

U.S. Shipments, Sales, Revenues and E-

Commerce: 2003

Value of Shipments, Sales or Revenue

($ billions) Description Total E-Commerce

Total 16,648 1,679

B-to-B* 8,296 1,573 Manufacturing 3,980 843 Merchant Wholesale 4,316 730

B-to-C* 8,352 106 Retail 3,275 56 Selected Services 5,077 50 Source: U.S. Census Bureau, E-stats: E-Commerce 2003

Highlights, May 2005. *The B-to-B and B-to-C breakdown was estimated by the Census Bureau and was not directly measured.

E-Stats Data, 2003

• 94% of e-commerce sales were B-to-B

• Only 1.7% of retail sales were e-commerce

• EDI (Electronic data interchange) is importantfor businesses– 86% of merchant wholesale e-commerce sales were

EDI

4

Retail E-Commerce is growing…

Estimated Quarterly U.S. Retail Sales: E-Commerce

0

5

10

15

20

25

1999

Q4

2000

Q1

Q2

Q3

Q4

2001

Q1

Q2

Q3

Q4

2002

Q1

Q2

Q3

Q4

2003

Q1

Q2

Q3

Q4

Q2

Q3

Q4

2005

Q1

r

Q2

p

Quarter

Bil

lio

ns

of

Do

lla

rs

Source: U.S. Census Bureau, August 2005.

…and replacing mail order sales

Estimated Annual Retail "Remote Sales" Using

Electronic Shopping and Mail-Order Houses in the United States

Billions of $

$76 $77

$88 $88

$79

$91

$4

$12

$21$25

$33

$40

$0

$40

$80

$120

$160

1998 1999 2000 2001 2002 2003

Bil

lio

ns

of

$

Mail Order E-commerce

Source: U.S. Census Bureau.

5

Compare Electronic “Retail Sales” -Shop.org and Census Bureau estimates

Comparison of Estimates for Electronic “Retail Sales” 2003

(Online) (E-commerce)

Retail Sales ($ Billions)

Annual Growth

Percent of Total Retail Sales

Shop.org, Forrester Research $114.1 51% 5.4%

Census Bureau E-Commerce $55.7 25% 1.7%

Source: Shop.Org data - Sacramento Bee, June 8, 2004 Section D. U.S. Census Bureau, E-Commerce Multi-Sector Historical Data Tables, 2003 Annual Retail Trade Survey.

Recent tax loss estimates45 states plus the District of Columbia

Compare Estimates: 2003 and 2008

$2.5

$15.5 $16.1

$3.7

$21.5

$33.7

$0.0

$5.0

$10.0

$15.0

$20.0

$25.0

$30.0

$35.0

$40.0

DMA Bruce/Fox low-growth Bruce/Fox high-growth

$ b

illi

on

s

2003

2008

45 states + District of Columbia

6

Tax Loss Estimates

• Calculate base sales– E-commerce sales (Internet, EDI, etc.)

• Taxability (exempt sales)• Compliance• Calculate sales resulting in loss• Apply state (+ local) tax rates

• Projections: e-commerce growth assumption

Taxability and compliance

• Consumer use tax compliance, very low• Most business transactions are not

subject to sales or use tax– Inputs to production processes– Purchases for resale

• AUDIT => (large) businesses…higher usetax compliance than small businesses orhouseholds

7

B-to-C E-Commerce

B-to-C: Compare E-Commerce Estimates

0

50

100

150

200

250

1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011

Bil

lio

ns

of

Do

lla

rs

Census B-to-C

Bruce/Fox Base

DMA Base

Bruce/Fox Sales Resulting in Loss

DMA Sales Resulting in Loss

DMA B-to-C Base Sales

Bruce/Fox B-to-C Sales Resulting in Loss

Bruce/Fox B-to-C Base Sales

DMA B-to-C Sales Resulting in Loss

Census

Data were adjusted to include only states with sales taxes.

B-to-B E-Commerce, lowB-to-B: Compare E-Commerce Estimates

Bruce/Fox Low-Growth Scenario Base

0

1,000

2,000

3,000

4,000

1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011

Bil

lio

ns

of

Do

lla

rs

Census B-to-B

Bruce/Fox Base LOW

DMA Base

Bruce/Fox Sales Resulting in Loss LOW

DMA Sales Resulting in Loss

Data were adjusted to include only states with sales taxes.

Bruce/Fox B-to-B Base Sales LOWCensus

DMA B-to-B Base Sales

Bruce/Fox B-to-B Sales Resulting in Loss LOW

DMA B-to-B Sales Resulting in Loss

8

B-to-B E-Commerce, highB-to-B: Compare E-Commerce Estimates

Bruce/Fox High-Growth Scenario Base

0

1,000

2,000

3,000

4,000

1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011

Bil

lio

ns

of

Do

lla

rs

Census B-to-B

Bruce/Fox Base HIGH

DMA Base

Bruce/Fox Sales Resulting in Loss HIGH

DMA Sales Resulting in Loss

Data were adjusted to include only states with sales taxes.

Bruce/Fox B-to-B Base Sales HIGH

Census

DMA B-to-B Base Sales

Bruce/Fox B-to-B Sales Resulting in Loss HIGH

DMA B-to-B Sales Resulting in Loss

Differences in Estimates:DMA - Bruce/Fox

Taxability and Compliance Assumptions

B-to-C: ServicesB-to-B: Adjustments to E-Commerce Data

for EDI, Interplant Sales and Services

Projected growth in E-commerce

9

Differences in estimates

Note: DMA estimates as published; Bruce/Fox breakdown as calculated by author.

2003 B2B and B2C Loss Estimates

$2.1

$0.4

$2.5$2.8

$12.7

$15.5

$0.0

$2.0

$4.0

$6.0

$8.0

$10.0

$12.0

$14.0

$16.0

$18.0

B2C B2B B2C+B2B

$ B

Illi

on

s

DMA

BRUCE/FOXlow-growth

States use different methods to estimate losses

• Use tax non-compliance (audit businesses)• E-commerce

– B2C– B2B

• B2B (Internet): DMA, E-stats data• B2B (E-commerce): Bruce/Fox, Forrester data

– State taxable retail sales; apply % of retail sales that are e-commerce

• Remote Sales (E-commerce + Mail Order)• B2C + Mail Order• B2C + B2B + Mail Order

• Sales and use tax losses– (losses from retail e-commerce + Mail Order)– + Other losses…use tax non-compliance rate for business

10

Independent State Estimates:E-commerce and Remote Sales

Except for Il l inois and Alabama, these are offic ial state estimates.

(State only) vs (State + Local)

*The Washington (02) remote sales estimate is for consumers only.

B2C B2B B2C+B2B Mail Order Remote Sales

CA (03) CA (03) CA (03) CA (03) CA (03)

MN (00) MN (00) MN (00) MN (00) MN (00)

MN (03)

IL (02) IL (02) IL (02)

PA (98) PA (98) PA (98)

PA (03)

MI (03) MI (03) MI (03)

WI (03) WI (03) WI (03)

WA (02)*

ME (03)

WV (02)

VT (03)

AL (03)

TX (03) TX (03) TX (03)

SC (05) SC (05) SC (05)

How do 2003 state estimates comparewith Bruce/Fox and DMA?

• Hard to compare…• Most states only estimate B2C and/or

mail order…• Alabama estimate is higher than

Bruce/Fox

11

Examples of state methods• Minnesota Tax Gap Study – American Economics Group -

databases of transaction and audit data…input-output approach

• Alabama – survey of middle-income consumers

• California – B2C + B2B – research on the B2B estimates

• Several States - Retail e-commerce as a % of taxable retailsales.

• Wisconsin, Michigan – Combine retail e-commerce/mail orderloss estimates with (business) use tax non-compliance rates

• Step 1: Estimate sales and use tax losseson retail e-commerce and mail order sales(consumers and business finalpurchases) …national Census data– State share of national retail e-commerce and mail order sales– Assumption about % with tax collected

• Step 2: Estimate other losses (business)

12

Step 2: other losses (business)

– Adjust collected use tax to exclude tax on retail sales and onoccasional sales of motor vehicles

– Calculate uncollected use tax using use tax compliance rate

Example:

• Collected use tax (adjusted) = $100m• Use tax compliance = 72%• Total use tax = $100m/0.72 = $139m• Uncollected use tax = $39m

For Minnesota, compare collected use taxvs. uncollected loss ($ million)

$133.9 m*Minnesota$50.9DMA

$331 m - $345 mBruce/FoxE-commerce2003 loss estimates

MN use tax revenue (2003) = $238.9 million

*Note: In 2000, 81% of the e-commerce gap was use tax; 19% sales tax.

E-commerce gap was all non-filers (not underreporting).

13

Compare collected vs. uncollected taxacross states

– Loss estimates– Actual use tax revenue– Estimated use tax revenue

Data Collection: 45 states with sales tax

112014*Lossestimate

10629Use tax

in progressnoyes

2 loss estimates (IL, AL) are not from official state sources.

14

Use tax definitions and data collectionvary across statesUse Tax as a Percentage of Sales and Use Tax Collections

Year: 2003 (for most states)

0%

2%

4%

6%

8%

10%

12%

14%

16%

18%

20%

Tenne

ssee

Neb

rask

a

Haw

aii

Florid

a

Ariz

ona

Wisco

nsin

Idah

o

Min

neso

ta

South

Car

olin

a

Was

hing

ton

Okl

ahom

a

Nor

th C

arol

ina

Texas

New

Jers

ey

Nor

th D

akot

a

Col

orad

o

Cal

iforn

ia

Mai

ne

Wes

t Virg

inia

South

Dak

ota

Wyo

min

g

Ala

bam

aIo

wa

Ohi

o

Ken

tuck

y

Kan

sas

Ark

ansa

s

Mic

higa

n

Louisia

na

Actual Use Tax Data compared with Loss Estimates: DMA,

BruceFox low, 2003

$0

$500

$1,000

$1,500

$2,000

$2,500

$3,000

ND WY SD ID ME WV NE IA KS AR OK SC AL CO WI MN LA AZ NJ WA MI OH FL TX CA

millio

ns o

f $

DMA

use

BF low

15

Actual Use Tax Data compared with Loss Estimates:

DMA, BruceFox low-growth, 2003

$0

$100

$200

$300

$400

$500

ND WY SD ID ME WV NE IA KS AR OK SC AL CO WI MN LA AZ NJ WA

mil

lio

ns

of

$

DMA

use

BF low

Adjusted state sales tax data toestimate use tax data across states

2003 Adjusted General State Sales & Gross Receipts Tax Collections ($ Millions)

1 ALABAMA* AL 1,894 23 MISSOURI MO 2,819

2 ARIZONA* AZ 4,287 24 NEBRASKA* NE 1,420

3 ARKANSAS AR 1,951 25 NEVADA NV 2,192

4 CALIFORNIA CA 24,899 26 NEW JERSEY NJ 5,936

5 COLORADO CO 1,833 27 NEW MEXICO* NM 1,491

6 CONNECTICUT CT 3,065 28 NEW YORK* NY 8,507

7 FLORIDA* FL 15,078 29 NORTH CAROLINA* NC 3,992

8 GEORGIA* GA 4,738 30 NORTH DAKOTA* ND 429

9 HAWAII* HI 1,707 31 OHIO OH 6,761

10 IDAHO ID 842 32 OKLAHOMA* OK 1,503

11 ILLINOIS* IL 6,613 33 PENNSYLVANIA PA 7,561

12 INDIANA IN 4,210 34 RHODE ISLAND* RI 766

13 IOWA IA 1,726 35 SOUTH CAROLINA* SC 2,576

14 KANSAS KS 1,888 36 SOUTH DAKOTA* SD 483

15 KENTUCKY* KY 2,820 37 TENNESSEE* TN 5,414

16 LOUISIANA* LA 2,776 38 TEXAS* TX 17,409

17 MAINE ME 857 39 UTAH UT 1,487

18 MARYLAND* MD 3,4605 40 VERMONT* VT 316

19 MASSACHUSETTS MA 3,708 41 VIRGINIA* VA 3,305

20 MICHIGAN* MI 7,652 42 WASHINGTON* WA 6,006

21 MINNESOTA* MN 4,904 43 WEST VIRGINIA* WV 1,139

22 MISSISSIPPI* MS 2,464 44 WISCONSIN* WI 3,741

45 WYOMING WY 425

Source: John L. Mikesell, State Tax Notes, July 2004.

*Asterisk denotes tax receipts were adjusted for comparability across states.

16

Compare estimated use tax toe-commerce losses, 20032003: Compare DMA, Bruce-Fox low-growth and Estimated State Use

Tax (7% of state SUT)

$0

$500

$1,000

$1,500

$2,000

$2,500

VT ND WY SD RI ID ME WV NE HI NM IA UT KS NV AR MS OK SC KY AL CT MD MA CO VA WI MO IN MN NC LA AZ NJ TN WA MI IL PA GA OH FL NY TX CA

DMA

estimated adjusted use tax (7% of SUT)

BF low-growth

Implications of states’ inability tocollect all use taxes due

• Loss of state and local government revenue.Losses are projected to increase due to growthin e-commerce.

• Firms have an incentive to locate productionand sales activity to avoid tax collectionresponsibility => efficiency losses

• Entity isolation• Equity: Sales tax becomes more regressive as

online sales grow if everyone doesn’t haveequal online access.

• Level playing field: Bricks vs Clicks

17

Bibliography• American Economics Group, Inc., Minnesota Sales and Use Tax Gap Project: Final Report, November 19, 2002.

http://www.taxes.state.mn.us/legal_policy/research_reports/content/tax_gap_study.shtml.• Bruce, Donald and Will iam F. Fox. State and Local Sales Tax Revenue Losses From E-Commerce: Estimates as of July 2004.

Knoxville, Tennessee: Center for Business and Economic Research, July 2004.http://www.nga.org/cda/fi les/0407ECOMMERCE.pdf http://cbaweb2a.bus.utk.edu/cber/ecomm/Ecom0704.pdf. Also publishedin State Tax Notes, August 16, 2004.

• California State Board of Equalization, Research and Statistics Section. Revenue Estimate: Electronic Commerce and MailOrder Sales, Draft to be Released.

• Civic Federation (The). The Potential Impact of the Streamlined Sales and Use Tax Agreement on Il l inois. Chicago, Il l inois: theFederation, October 18, 2004. http://www.civicfed.org/artic les/civicfed_157.pdf.

• Haas, Mark P. and Andrew Lockwood, Tax Revenue Loss Estimates for Consumer Remote Sales, Michigan Department ofTreasury, April 2001.

• Johnson, Peter A. A Current Calculation of Uncollected Sales Tax Arising From Internet Growth. New York: Direct MarketingAssociation, March 11, 2003. http://www.thedma.org/taxation/CurrentCalculationofUncollectedSalesTax.pdf.

• Jones, Martha. The Streamlined Sales and Use Tax Agreement: A California Perspective, California Research Bureau, February2005. http://www.library.ca.gov/crb/05/01/05-001.pdf http://www.library.ca.gov/html/statseg2a.cfm

• Michigan Department of Treasury, Bureau of Tax and Economic Policy, Tax Analysis Division. Michigan’s Sales and Use Taxes2004, June 2005, Pages 8, 25-34. http://www.michigan.gov/documents/SUWReport_June2005_129823_7.pdf

• Mikesell, John L. “State Retail Sales Tax Burdens, Reliance, and Breadth in Fiscal 2003.” State Tax Notes, July 12, 2004.• Pennsylvania Department of Revenue. “The Impact of Electronic Commerce on Pennsylvania Sales and Use Tax: February

2000 Update.” Harrisburg, PA: The Department, February 2000.http://www.revenue.state.pa.us/revenue/lib/revenue/ecommerceimpact.pdf.

• Singleton, Tommie W., Brett A King, and Aaron J. Singleton. “Internet Sales Tax: A Survey of Revenues Lost to Alabama,” StateTax Notes, January 10, 2005.http://www.business.uab.edu/FACULTY/TSingleton/Research/State%20Tax%20Notes/State.Tax.Notes-Internet.Sales.Tax.pdf

• United States General Accounting Office. Sales Taxes: Electronic Commerce Growth Presents Challenges; Revenue Lossesare Uncertain, June 30, 2000. GAO/GGD/OCE-00-165.

• United States General Accounting Office. Update on State and Local Revenue Loss From Internet Sales, November 6, 2001.GAO-02-83R.

• Washington State Department of Revenue. Department of Revenue Compliance Study, Olympia, Washington: The Department,January 17, 2003. http://dor.wa.gov/Docs/Reports/Compliance_Study/compliance_study_2003.pdf,http://dor.wa.gov/docs/pubs/news/2003/nr_03_01_%20usetxnoncompliance.pdf

• Wisconsin Department of Revenue, Division of Research and Policy. Estimated Fiscal Effect of E-Commerce and Other RemoteSales, July 3, 2003.

Contact information

Martha JonesCalifornia Research Bureau, California

State Library, Sacramento, California(916) 653-6742mjones@library.ca.gov