Presentation for the Niagara University Conference - April 2008

28
Bay is a trademark of eBay corporation Why eBay? In QIII of 2007 1. Number of listings was around 556 million 2. Value of successfully sold merchandise was around $14.4 billion 3. Confirmed registered users: 247.6 million 4. Active users: 83 million For comparison: According to the US Department of the Census During the third quarter of 2007: Total online retail commerce was about $34.7 billion (generated by some 12,500 online businesses)

Transcript of Presentation for the Niagara University Conference - April 2008

Page 1: Presentation for the Niagara University Conference - April 2008

eBay is a trademark of eBay corporation

Why eBay?

In QIII of 20071. Number of listings was around 556 million

2. Value of successfully sold merchandise was around $14.4 billion3. Confirmed registered users: 247.6 million

4. Active users: 83 million

For comparison:According to the US Department of the Census

During the third quarter of 2007:Total online retail commerce was about $34.7 billion

(generated by some 12,500 online businesses)

Page 2: Presentation for the Niagara University Conference - April 2008

Price Distribution and the Choice of Promotion

• Key difference: introduction of price distribution characteristics

• multi-product study– Distributions differ in terms of

• Mean realized prices• Standard deviation• Kurtosis/skewdness• Available on eBay supply• Difference in eBay valuation and the retail price

• Larger dataset – Still GPS devices (multiple models)

Page 3: Presentation for the Niagara University Conference - April 2008

Why GPS?• Restricted to brand new items only

– Completely homogeneous good– Likely to observe homogenized shipping

methods

• Relatively popular on eBay category– GPS auto devices constitute an independent

category• During our data collection period nearly 3000

listings for brand new (non-refurbished) GPS devices were available on eBay

• Price comparison with retail websites

Page 4: Presentation for the Niagara University Conference - April 2008

GPS models in our dataset

GPS model Obs. Obs.

Multi-Item

ListingsAverage

PriceAverage

PriceSt. Dev. of

Price Price Range

(total)(sold items only)

(number of

listings)

(all listings)

(sold items only)*

(sold items only)

(sold items only)

GARMIN200W 188 112 63 221.16 208.54 17.74 159.99 - 274.99

260 87 69 21 247.51 242.72 19.66 165.60 - 304.99360 156 85 40 328.25 306.30 51.79 232.50 - 472.00370 16 9 10 456.15 421.77 37.83 369.99 - 479.99660 180 71 53 436.30 404.19 38.66 250.00 - 497.00750 62 38 32 413.30 397.95 38.39 259.99 - 459.95760 134 89 44 524.33 509.63 27.93 425.00 - 589.99770 39 20 24 674.08 648.78 41.76 560.00 - 699.99

C330 43 35 0 146.74 141.60 20.19 89.99 - 177.50C340 98 50 25 212.03 198.99 24.21 125.00 - 229.99C530 63 22 13 213.56 197.04 24.17 157.50 - 229.99C550 17 6 12 344.28 331.62 12.97 309.99 - 339.99

MAGELLAN3140 30 21 4 213.14 208.84 24.94 170.00 - 274.854040 48 35 8 276.18 248.85 33.60 132.49 - 290.004050 27 12 7 350.99 316.75 36.71 255.00 - 392.99

TOMTOM720 131 108 13 324.10 315.94 25.84 152.51 - 379.99920 43 36 7 441.36 434.29 25.12 389.99 - 499.99

HPiPAQ 310 71 58 23 241.58 235.53 18.01 189.10 - 289.99

MIOC520 15 9 6 250.84 233.65 24.24 202.50 - 279.95

Page 5: Presentation for the Niagara University Conference - April 2008

Feature Description Fee

Featured In addition to being displayed in the search results, the listing will also be displayed in the top section of the search page. This form of promotion is only available to sellers whose rating is at least

10.

19.95

GalleryPlus

Creates a pop-up window displaying a larger gallery image file when a mouse pointer is moved

over the listing on the search page.

0.75

Gallery

Enables the seller to display a small immage on the search page next to the title of the listing.

0.35

Highlight

Provides a color background to the title of the listing on the search page, effectively highlighting

the listing on the search page.

5.00

Border

Provides a color border around the listing on the search page.

3.00

Bold

The title of the listing on the search page is in bold font.

1.00

Subtitle

The seller is allowed to enter an additional line of text to be displayed on the search page below the

title of the auction.

0.50

Page 6: Presentation for the Niagara University Conference - April 2008

Multi-item listings onlyGPS model Listings

Average St.Dev Range Average St.Dev Range

GARMIN200W 63 9.556 11.814 2 - 55 1.635 5.182 0 - 37

260 21 22.619 17.060 2 - 49 3.857 4.004 0 - 14360 40 11.500 11.715 2 - 40 1.000 2.602 0 - 11370 8 27.000 16.844 2 - 40 1.000 2.070 0 - 6660 53 7.698 6.062 2 - 20 0.358 0.710 0 - 4750 32 14.469 12.051 2 - 40 1.625 2.992 0 - 16760 44 13.977 12.199 2 - 40 1.432 2.840 0 - 15770 24 11.083 6.107 2 - 20 0.458 0.721 0 - 2

C330 0C340 25 17.000 15.324 2 - 50 2.520 3.721 0 - 14C530 13 15.692 17.447 2 - 40 1.769 3.632 0 - 13C550 12 16.583 12.944 2 - 30 1.167 1.586 0 - 5

MAGELLAN3140 4 10.000 0.000 10 1.750 2.217 0 - 54040 8 16.250 3.284 10 - 18 2.000 3.071 0 - 84050 7 14.714 5.880 4 - 20 0.000 0.000 0

TOMTOM720 13 12.769 9.791 2 - 30 3.231 5.036 0 - 18920 6 26.000 31.956 7 - 90 7.333 8.869 1 - 25

HPiPAQ 310 21 5.048 5.599 2 - 27 1.143 1.014 0 - 3

MIOC520 6 7.833 6.735 2 - 20 1.000 2.000 0 - 5

All Items 400 12.703 12.596 2 - 90 1.540 3.467 0 - 37

Items per Listing Succesful Completion

Page 7: Presentation for the Niagara University Conference - April 2008

GPS model Obs. Featured GalleryPlus Gallery Highlight Border Bold Subtitle

GARMIN200W 188 0.0585 0.0266 0.9468 0.0372 0.0426 0.1223 0.5426

260 87 0.1264 0.0575 0.9310 0.0000 0.0000 0.0805 0.5517360 156 0.0577 0.0192 0.8462 0.0128 0.0192 0.0897 1.0000370 16 0.3125 0.1250 0.9375 0.1250 0.1250 0.4375 0.6250660 180 0.0833 0.0111 0.9167 0.0278 0.0222 0.0611 0.9944750 62 0.1774 0.0806 0.9355 0.0161 0.0161 0.2419 0.5323760 134 0.1045 0.0075 0.8284 0.0448 0.0448 0.1045 1.0000770 39 0.1282 0.0256 0.7179 0.0513 0.1795 0.1282 1.0000

C330 43 0.0000 0.0000 0.8140 0.0233 0.0000 0.0000 0.0698C340 98 0.1122 0.0306 0.9286 0.0408 0.0102 0.0816 0.5204C530 63 0.0952 0.0000 0.9683 0.0159 0.0635 0.1270 0.6667C550 17 0.2941 0.0000 1.0000 0.0000 0.0000 0.4706 0.7647

MAGELLAN3140 30 0.1333 0.0000 0.8333 0.0000 0.0000 0.0000 0.50004040 48 0.0625 0.0000 0.9375 0.0000 0.0000 0.1875 0.37504050 27 0.1111 0.0370 0.9630 0.0370 0.0741 0.2593 0.8148

TOMTOM720 131 0.0611 0.0611 0.9084 0.0229 0.0458 0.1527 0.4122920 43 0.0930 0.0233 0.8837 0.0233 0.0465 0.3256 0.4186

HPiPAQ 310 71 0.0563 0.0141 0.8592 0.0282 0.0000 0.0282 0.3662

MIOC520 15 0.0667 0.0000 0.8000 0.0667 0.0000 0.0000 0.6667

ALL 1448 0.0898 0.0262 0.8964 0.0269 0.0318 0.1188 0.6720

Single-Item Only 1048 0.0057 0.0191 0.8884 0.0076 0.0105 0.0697 0.6078

Multi-Item Only 400 0.3100 0.0450 0.9175 0.0775 0.0875 0.2475 0.8372

Average Values for Binary Variables

Occurrence of promotional tools in our dataset

Page 8: Presentation for the Niagara University Conference - April 2008

Retail Price comparison

Price

BestBuy.com

OfficeDepot.com

OfficeM

ax.com

CircuitCity.com

Manufacturer's Website

GPS model

DATA Average

price(sold items

only)

GARMIN200W 208.54 321.41 299.99 269.99

260 242.72 374.99 349.99 349.99 319.99360 306.30 499.99370 421.77 599.99 599.99 599.99660 404.19 749.99 499.99 649.99 899.99750 397.95 589.27 549.99 599.99760 509.63 749.99 699.99 799.99 699.99770 648.78 964.27 999.99

C330 141.60 299.99 129.99C340 198.99 299.99 199.99C530 197.04 299.99C550 331.62 482.13 449.99 389.99 349.99

MAGELLAN3140 208.84 349.99 249.994040 248.85 449.99 299.99 499.994050 316.75 499.99 499.99 699.99 399.99

TOMTOM720 315.94 449.95 449.99 449.99 419.99920 434.29 549.95 549.99

HPiPAQ 310 235.53 399.99 399.99 429.99

MIOC520 233.65 399.99 249.99

BestBuy.com

OfficeDepot.com

OfficeM

ax.com

CircuitCity.com

Manufacturer's Website

Page 9: Presentation for the Niagara University Conference - April 2008

Overlap in the use of promotional tools(count of observations)

Featured 130

GalleryPlus 16 38

Gallery 130 38 1298

Highlight 31 16 37 39

Border 29 17 46 20 46

Bold 75 27 170 19 34 172

Subtitle 128 26 895 35 44 147 973

Page 10: Presentation for the Niagara University Conference - April 2008

Choice of Promotional tools

Probit Estimation

Variableb dY/dX b dY/dX b dY/dX

Pmean -0.0001 -6.30E-06 0.0007 0.0001 0.0014 0.0005(0.22) (1.70) (3.77)

St. Dev. -0.004 -0.00019 -0.0036 -0.0007 0.0241 0.009(0.58) (0.72) (5.76)

Opening2mean -0.1387 -0.00666 -0.2184 -0.0421 0.4385 0.1638(0.25) (0.66) (1.71)

Frequency -0.0021 -0.0001 -0.0023 -0.0005 0.004 0.0015(1.96) (2.98) (6.07)

Multi-item 1.8862 0.235364 0.8204 0.1976 0.3697 0.1331(10.42) (7.87) (3.90)

ln (Rating+1) 0.1016 0.004874 0.0147 0.0028 0.1097 0.041(2.68) (0.64) (6.32)

Rating_Per 0.2439 0.011707 0.1641 0.0316 0.0617 0.0231(2.29) (2.71) (2.36)

Shipping -0.0001 -6.52E-06 -0.0124 -0.0024 0.0111 0.0042(0.02) (1.87) (2.04)

FriSatSun -0.0882 -0.00418 -0.0646 -0.0124 0.1573 0.0585(0.70) (0.72) (2.11)

6 pm - 11 pm 0.0505 0.002472 -0.0651 -0.0124 0.0645 0.024(0.35) (0.65) (0.80)

Constant -26.915 -17.217 -8.9335(2.52) (2.84) (3.42)

Pseudo-R2 0.3699 0.1134 0.1648

Observations 1413 1413 1413

Featured HBB GS

Interesting points1. Mean value increases

use of most promotional tools except for FEATURED (a very expensive tool)

2. Frequency reduces the need for visual promotion but increases the usage of informational tools

3. Experience and reliability seem to increase the demand for promotional tools

4. Multi-items are much more likely to use promotional tools, largely due to economies of scale

Page 11: Presentation for the Niagara University Conference - April 2008

Probit Estimation

Choice of Promotional

tools II

Variableb dY/dX b dY/dX b dY/dX b dY/dX

Pmean 0.003011 0.000954 0.000665 0.000115 0.0016305 0.0000704 0.000644 0.0000216(6.23) (1.52) (2.55) (0.90)

St. Dev. 0.045066 0.014607 -0.002814 -0.000487 -0.0012487 -0.0000539 -0.006101 -0.0002044(8.00) (0.56) (0.15) (0.69)

Opening2mean 0.865445 0.225258 -0.161087 -0.027865 -0.8584872 -0.0370818 -0.83796 -0.0280701(3.07) (0.47) (1.31) (1.24)

Frequency 0.00554 0.001777 -0.00203 -0.000351 0.0000418 1.81E-06 0.000748 0.0000251(7.51) (2.49) (0.03) (0.55)

Multi-item 0.3857 0.128553 0.720889 0.151669 1.094942 0.0850229 1.160504 0.0763521(3.57) (6.47) (5.49) (5.35)

ln (Rating+1) 0.133443 0.043956 0.019418 0.003359 -0.0155065 -0.0006698 -0.028648 -0.0009597(6.89) (0.82) (0.38) (0.67)

Rating_Per 0.027759 0.008713 0.145922 0.025242 -0.0109029 -0.0004709 0.190692 0.0063878(1.18) (2.32) (0.28) (1.39)

Shipping 0.009792 0.002957 -0.01122 -0.001941 -0.0217646 -0.0009401 -0.00707 -0.0002368(1.58) (1.64) (2.16) (0.66)

FriSatSun 0.130912 0.040545 -0.016256 -0.002808 -0.1341547 -0.0057008 -0.107058 -0.0035362(1.59) (0.18) (0.89) (0.67)

6 pm - 11 pm -0.041814 -0.016004 -0.168034 -0.027884 0.1218971 0.0055428 -0.082989 -0.0026854(0.48) (1.59) (0.75) (0.46)

Constant -7.002731 -15.60692 -0.4874321 -20.3715(2.97) (2.48) (0.13) (1.49)

Pseudo-R2 0.2877 0.0975 0.1547 0.1493

Observations 1413 1413 1413 1413

Subtitle Bold Border Highlighted

Page 12: Presentation for the Niagara University Conference - April 2008

Choice of Promotional tools

in multi-item listings

Variableb dY/dX b dY/dX b dY/dX

Pmean -0.000404 -0.000129 -0.000104 -0.0000337 0.001811 0.000289(0.48) (0.14) (2.15)

St. Dev. 0.005569 0.00177 0.000193 0.0000623 0.002026 0.000323(0.55) (0.02) (0.24)

Opening2mean 0.790359 0.251204 0.794105 0.2565379 1.749508 0.27919(0.93) (1.06) (2.27)

Frequency 0.001247 0.000396 -0.000717 -0.0002317 0.001161 0.000185(0.78) (0.52) (0.78)

Items_in_Listing 0.110541 0.035134 0.054966 0.0177569 0.108468 0.01731(10.84) (8.10) (5.79)

ln (Rating+1) 0.193137 0.061386 0.1384 0.0447105 0.08152 0.013009(3.39) (2.77) (1.63)

Rating_Per 0.740703 0.235422 0.652388 0.2107557 0.140595 0.022437(4.48) (5.06) (1.11)

Shipping -0.036788 -0.011692 -0.011951 -0.0038609 -0.006072 -0.000969(2.83) (1.04) (0.58)

FriSatSun -0.164738 -0.051655 -0.035604 -0.0114708 0.24813 0.03817(0.89) (0.23) (1.45)

6 pm - 11 pm 0.267289 0.088683 -0.121456 -0.0384001 0.0359 0.005657(1.22) (0.63) (0.18)

Constant -77.80323 -67.91792 -17.43532(4.64) (5.24) (1.38)

Pseudo-R2 0.4672 0.2634 0.2418

Observations 400 400 400

Featured HBB GS

1. Mean price impacts informational tool usage

2. Economies of scale effect is seen in the listing item effect

Page 13: Presentation for the Niagara University Conference - April 2008

Choice of Promotional tools in multi-item listings II

Variableb dY/dX b dY/dX b dY/dX b dY/dX

Pmean 0.004629 0.000334 -0.000301 -0.0000762 0.00172 0.000218 0.000532 0.000053(3.47) (0.39) (1.96) (0.53)

St. Dev. 0.019078 0.001375 0.003985 0.0010078 0.004043 0.000513 -0.002127 -0.000212(1.66) (0.43) (0.37) (0.18)

Opening2mean 2.361131 0.170114 0.726531 0.1837612 -1.229725 -0.156059 -0.291584 -0.029046(2.69) (0.91) (1.04) (0.25)

Frequency 0.002223 0.00016 -0.001248 -0.0003157 0.002793 3.54E-04 0.004881 0.000486(1.32) (0.83) (1.54) (2.45)

Items_in_Listing 0.112327 0.008093 0.054711 0.0138379 0.029695 0.003768 0.04372 0.004355(4.58) (7.66) (4.07) (5.45)

ln (Rating+1) 0.163558 0.011784 0.202731 0.0512767 0.034816 0.004418 -0.099626 -0.009924(2.80) (3.59) (0.54) (1.42)

Rating_Per 0.083209 0.005995 0.898061 0.2271462 0.115983 0.014719 0.245153 0.02442(0.62) (5.00) (0.87) (1.48)

Shipping -0.019814 -0.001428 -0.003226 -0.0008159 -0.036859 -0.004678 -0.014675 -0.001462(1.63) (0.26) (2.58) (0.93)

FriSatSun 0.311809 0.021227 0.04715 0.0119872 -0.304056 -0.036771 -0.141923 -0.013778(1.52) (0.28) (1.47) (0.64)

6 pm - 11 pm -0.215739 -0.017307 -0.636051 -0.1358286 0.229252 0.031887 0.163516 0.017511(1.00) (2.77) (0.97) (0.65)

Constant -13.96268 -93.14749 -12.72044 -25.91494(1.04) (5.17) (0.95) (1.56)

Pseudo-R2 0.3787 0.3242 0.1211 0.1861

Observations 400 400 400 400

Subtitle Bold Border Highlighted

Page 14: Presentation for the Niagara University Conference - April 2008

Impact of promotional tools on the probability

of saleProbit model

b dY/dX b dY/dX b dY/dX b dY/dX

Subtitle -0.06876 -0.0192 -0.0511 -0.0143(0.52) (0.39)

HBB 0.327271 0.081209 0.3404 0.0840(1.50) (1.56)

Gallery 0.16841 0.049873 0.1569 0.0464 0.1751 0.0520(0.93) (0.88) (0.98)

Opening2mean -10.0317 -2.81914 -10.0550 -2.8296 -10.0866 -2.8369 -10.0542 -2.8246(16.16) (16.45) (16.35) (16.24)

Frequency -0.00379 -0.00107 -0.0040 -0.0011 -0.0040 -0.0011 -0.0037 -0.0011(3.39) (3.76) (3.72) (3.36)

ln (Rating+1) -0.03827 -0.01075 -0.0453 -0.0128 -0.0423 -0.0119 -0.0382 -0.0107(1.61) (1.97) (1.83) (1.62)

Rating_Per 0.039593 0.011126 0.0421 0.0118 0.0402 0.0113 0.0424 0.0119(1.65) (1.76) (1.67) (1.77)

Shipping -0.01639 -0.00461 -0.0171 -0.0048 -0.0184 -0.0052 -0.0162 -0.0046(2.04) (2.17) (2.33) (2.02)

FriSatSun -0.10343 -0.02918 -0.1036 -0.0293 -0.1062 -0.0300 -0.0968 -0.0273(0.94) (0.95) (0.97) (0.89)

6 pm - 11 pm -0.10139 -0.02889 -0.0754 -0.0214 -0.0885 -0.0252 -0.0971 -0.0276(0.88) (0.66) (0.77) (0.84)

Retail_Price 0.0010 0.0003(2.66)

Retail-Pmean 0.002834 0.000797 0.0028 0.0008 0.0026 0.0007(3.08) (3.25) (2.96)

Constant 7.215013 6.9573 7.2590 7.1027(3.03) (2.94) (3.05) (2.97)

Pseudo-R2 0.4543 0.4495 0.4524 0.4536

Observations 1013 1013 1013 1013

Y = Sold Y = Sold Y = Sold Y = Sold

1. Promotion seems to leave the probability of sale largely unaffected

2. Price characteristics play expected roles1. Price deviation from

the mean3. Shipping negatively

impacts sales4. Positive rating has a

robust impact on the probability of sale

5. Supply has a negative effect on the probability of sale (frequency)

Page 15: Presentation for the Niagara University Conference - April 2008

Promotional tools and willingness to pay

Tobit model with varying censoring point Amemya Variation

Variable Coefficient(t-stat)

Featured 17.26136(0.84)

HBB 10.91526(1.79)

GS -1.896295(0.50)

Frequency -0.041513(1.25)

ln (Rating+1) -0.2054002(0.28)

Rating_Per 0.7888234(1.16)

Shipping -1.000835(3.86)

FriSatSun -5.224928(1.54)

6 pm - 11 pm 1.831352(0.51)

Retail_Price 0.6335644(57.62)

Constant -14.95922(0.22)

Pseudo-R2 0.1639

Observations 1013

Interesting points:1. HBB seems to increase the willingness to

pay by about $10.92 ( about 3% of the average price, which in our dataset is $331.65)

2. Shipping reduces willingness to pay one for one

3. Retail price increases the willingness to pay by 63.4% for every dollar [around the average values]

Page 16: Presentation for the Niagara University Conference - April 2008

Future directions

• Investigate the rating impact– Role of positive seller rating on the auction outcome

• Increase dataset size by including high frequency items– A study on a single high frequency item is a must to

fully investigate the role of visual promotional tools HBB, as these should be expected to play an increasingly more important role

• Price comparison with retail websites needs to be further investigated empirically. – Price dynamics

• Can price trends on eBay predict discounts in real and online commerce (BestBuy….)?

Page 17: Presentation for the Niagara University Conference - April 2008

eBay is a trademark of eBay corporation

Why eBay?

In QIII of 20071. Number of listings was around 556 million

2. Value of successfully sold merchandise was around $14.4 billion3. Confirmed registered users: 247.6 million

4. Active users: 83 million

For comparison:According to the US Department of the Census

During the third quarter of 2007:Total online retail commerce was about $34.7 billion

(generated by some 12,500 online businesses)

Page 18: Presentation for the Niagara University Conference - April 2008

Insertion Fees and Their Role on eBay Mikhail I. Melnik

Niagara University

Success of an online auction website is in its appeal to the bidders. Without the bidders there will be no transactions. In an auction, the reservation price can act as a deterrent to bidder entry. The auctioneer can induce lower reservation prices through a rising insertion fee scheme. This will in turn increase the probability of an auction resulting in a sale. This work examines the impact the insertion fees have on the seller and buyer behavior on eBay using data generated by a natural experiment observed on eBay. The results demonstrate that the rising insertion fee structure of eBay induces sellers to lower their reservation prices, which in turn increases the bidding activity.

Page 19: Presentation for the Niagara University Conference - April 2008

Neglect of Insertion fees in economic literature

• Reservation pricing received substantial attention in the literature– Myerson (1981) demonstrated that selection of the opening price

can impact the revenues to the seller – Riley and Samuelson (1981) provided theoretical derivations of

optimal reserve price strategies for auctions with private values and risk neutral bidders.

– Levin and Smith (1996) extended the analysis to common value goods

– Robinson (1995) provided a theoretical formulation for cases with collusive behavior

• However, the existing research tends to focus on the problem of the seller, and tends to ignore the problem of the auctioneer, i. e. the selection of the insertion fee scheme

Page 20: Presentation for the Niagara University Conference - April 2008

Impact of Insertion Fees on the Seller’s Behavior

Characteristics of the fees

1. non-refundable

2. Independent of the auction outcome

3. Function of the opening auction price (reservation price)

Expected Impact

Seller Utility Maximization Problem

Melnik, Xu, Alm (2006) derive several propositions for the optimization problem above (the problem is simplified to income flow maximization under uncertainty)

Propositions

1. Opening price < reservation value (due to the non-refundable nature of the fee)

2. Opening price decreases as the marginal fee increases

max

0

)())(,()())(,( 00

P

P

o PPCUdPPPCPUEU

Page 21: Presentation for the Niagara University Conference - April 2008

Motivation for This ResearchCompetition Between eBay and Yahoo

• eBay is established on September 5, 1995• Yahoo Auctions is launched on September 14, 1998

– eBay user base is less than 1 million– In September of 1999, Yahoo user base was estimated (for all

segments of the business) to be around 100 million– Yahoo attempts to attract eBay sellers and offers zero insertion fee

scheme• In just one year, September 1999, the average daily number of

listings on Yahoo reaches 725,000 but the proportion closing successfully is very low

• To avoid seller demoralization Yahoo deletes from its archive site all non-successfully completed listings the moment they end.

• In 2001 Yahoo introduces insertion fees of 0.20 (considerably lower the eBay fee structure) with various sales available to its sellers. The fee structure is abandoned three years later

• In the summer of 2007 Yahoo suspends its US Auction website

Page 22: Presentation for the Niagara University Conference - April 2008

Insertion Fee Structure on eBayStarting or Reserve Price

Insertion Fee

$0.01 - $0.99 $0.20

$1.00 - $9.99 $0.40

$10.00 - $24.99 $0.60

$25.00 - $49.99 $1.20

$50.00 - $199.99

$2.40

$200.00 - $499.99

$3.60

$500.00 or more

$4.80

0.0

1.0

2.0

3.0

4D

ensi

ty

0 100 200 300 400 500Starting_price

0.0

02

.00

4.0

06

.00

8.0

1D

ensi

ty

0 100 200 300 400 500Starting_price0

.05

.1.1

5.2

.25

Den

sity

0 20 40 60Starting_price

Other days<500

March 14<500

Steps <60

Page 23: Presentation for the Niagara University Conference - April 2008

Data• Natural Experiment: March 14, 2007

– eBay market research– Unannounced change in the fee structure to $0.20

regardless of the opening price– One day only

• Selecting the item for such study– Opening price is a function of price distribution

• Multiple distributions– Controlling for item specific characteristics and

reducing the noise in the data– Popularity on eBay

• Generating sufficient data over a short period of time (1 day)– Morgan Dollars in AU condition

Page 24: Presentation for the Niagara University Conference - April 2008

Descriptive Statistics – Single Item Listingsvariable mean Std. dev min max mean Std. dev min max

ClosingPrice 87.109 141.301 0.01 999.99 163.728 175.368 0.99 999.99

ClosingPrice (sold coins) 55.656 87.741 5.12 999 71.225 90 9.99 510

ClosingPrice* (sold coins) 54.661 87.529 5.12 999

OpeningPrice 62.131 138.489 0.01 999.99 150.153 178.752 0.01 999.99

OpeningPrice* 47.879 125.172 0.01 999.99

Pmean 72.899 105.931 16.871 999 96.881 94.416 16.871 714.997

Std_Dev 33.398 37.267 0 225.676 46.027 38.958 0 147.569

CoinCount 41.007 28.13 1 131 37.962 21.671 1 131

ln(Rating+1) 7.433 1.695 0.693 10.816 7.848 1.509 2.944 10.758

Rating_percent 99.788 0.347 96.7 100 99.771 0.363 97.4 100

Bids 5.085 5.306 0 41 2.17 4.879 0 41

Certified 0.078 0 1 0.142 0 1

Completed 0.746 0 1 0.322 0 1

Weekend 0.308 0 1

FeeAlternative 0.143 0 1

Number of observations 2318 323

All Observations Auctions Starting on March 14

Page 25: Presentation for the Niagara University Conference - April 2008

Descriptive Statistics – Multi-item listings

variable mean Std. dev min max mean Std. dev min max

ClosingPrice 183.312 160.533 0.01 1499 215.405 116.591 20.95 592

ClosingPrice (sold coins) 114.573 122.484 14.25 549.99 121.063 79.252 20.95 295

OpeningPrice 157.055 166.549 0.01 1499 214.739 117.638 0.99 592

ln(Rating+1) 8.513 1.772 0.693 9.59 9.164 1.455 2.944 9.589

RatingPercent 99.822 0.16 98.3 100 99.798 0.083 99.4 100

Bids 3.107 6.145 0 34 0.15 0.702 0 6

Certified 0 0 0 0 0 0

Completed 0.414 0 1 0.08 0 1

Weekend 0.215 0 1

FeeAlternative 0.383 0 1

Number of observations 261 100

All Observations Auctions Starting on March 14

Page 26: Presentation for the Niagara University Conference - April 2008

Number of listings - OLS regression

with robust standard errors variable I II III IV V VI

FeeAlternative 238.173*** 249.114*** 232.644*** 233.439*** 230.723*** 85.08***(0.558) (0.471) (0.636) (0.760) (1.133) (1.003)

Weekend 31.658***(1.034)

Monday -57.824*** -58.032*** -57.415*** -1.354(0.861) (1.073) (1.444) (1.243)

Tuesday -34.246*** -33.722*** -35.378*** -6.142***(0.709) (0.879) (1.185) (1.133)

Wednesday -21.424*** -22.066*** -19.779*** -0.606(0.747) (0.903) (1.313) (1.427)

Thursday -41.987*** -41.836*** -42.256*** -11.617***(0.899) (1.111) (1.532) (1.060)

Friday -36.229*** -36.940*** -33.579*** -5.115***(0.978) (1.119) (2.036) (1.533)

Saturday -15.013*** -14.551*** -16.251*** -7.288***(2.001) (2.299) (4.099) (1.210)

Constant 108.827*** 120.479 135.78*** 135.626*** 136.056*** 15.526***(0.557) (93.865) (0.392) (0.487) (0.664) (1.015)

N 2349 2349 2349 1548 801 267

R20.928 0.955 0.966 0.945 0.982 0.993

Specifications I – V are limited to the single coin auctions only, while specification VI examines the multi-coin auctions offered in the AU Morgan Dollar category. Specification IV is restricted to the coins with Pmean of 50 dollars and lower, while Specification V is estimated on auctions of the coins with Pmean exceeding 50 dollars.

Page 27: Presentation for the Niagara University Conference - April 2008

Impact on opening price – OLS with robust standard errors

variable I II III IV V VI VII IV V

FeeAlternative 77.548*** 63.509*** 82.859*** 27.003*** 51.187** 79.045*** 129.392*** 89.571* 6.949

(8.465) (8.123) (13.849) (7.106) (21.615) (13.700) (49.418) (51.986) (5.449)

Weekend -20.552*** -53.528*** -1.536 -15.578*** -57.572*** -41.108 -95.255** -1.175

(4.081) (12.431) (2.132) (5.239) (10.033) (60.250) (39.889) (2.919)

ln(Rating+1) 2.405** 11.733*** 0.985 -4.178** 5.448** 48.957*** 7.283 3.15

(1.183) (3.104) (0.784) (1.765) (2.780) (15.222) (9.535) (2.160)

RatingPercent 6.841 0.881 6.658*** 19.538*** 16.043 -161.597 -16.791 2.307

(4.289) (20.151) (1.410) (6.870) (17.878) (162.660) (52.980) (1.879)

Certified 68.168*** 85.873*** 27.68 44.524** 66.768*** 151.935*** 254.829**

(12.645) (17.050) (19.196) (20.595) (15.031) (50.293) (108.410)

Pmean 0.858*** 0.654*** 0.552*** -1.308 1.136 0.639*** 0.324

(0.066) (0.122) (0.161) (0.890) (0.709) (0.092) (0.401)

Std_Dev 0.445** 1.006** 0.092 0.658 1.032*** 1.338

(0.207) (0.399) (0.215) (0.453) (0.374) (1.112)

CoinCount -0.063 -0.616 -0.068 -0.275 -0.638 -5.249

(0.090) (0.506) (0.084) (0.468) (0.537) (3.243)

Constant -8.604** -704.83* -191.744 -629.32*** -1940.59*** -1660.526 1582.529 1773.596 -240.229

(3.565) (426.076) (2012.728) (132.669) (687.673) 1781.168 1622.222 (5321.994) (189.096)

N 2305 2305 758 1088 459 639 119 85 131

R20.403 0.436 0.383 0.061 0.115 0.282 0.341 0.215 0.038

Specifications I and II are performed on the entire single-item dataset. Specification III is restricted to the coins with Pmean of 50 dollars and higher (the relatively expensive Morgan dollar coins). Specifications IV – VII are matched to the insertion fee structure of eBay in terms of the selection of the coins based on Pmean.

Page 28: Presentation for the Niagara University Conference - April 2008

Impact on the probability of sale

variable dy/dx dy/dx dy/dx dy/dx

OpeningBid/Pmean -1.174*** -0.381 -0.929** -0.247

-0.352 -0.364

OpeningBid -0.005** -0.002 -0.012* -0.003

-0.002 -0.007

Weekend 0.188** 0.059 0.086 0.023 0.833*** 0.304 -0.502* -0.12

-0.089 -0.087 -0.316 -0.283

ln(Rating+1) 0.004 0.001 0.051** 0.014 0.017 0.006 0.082 0.018

-0.023 -0.025 -0.092 -0.106

RatingPercent 0.048 0.016 -0.046 -0.012 -1.083 -0.421 0.558 0.121

-0.098 -0.116 -0.794 -0.343

Certified 0.017 0.006 0.164 0.041 0.428 0.155

-0.221 -0.21 -0.889

Shipping -0.091*** -0.024

-0.029

Constant -3.245 6.083 108.551 -54.822

-9.702 -11.527 -79.369 -34.303

N 2318 1606 85 130

Pseudo R20.357 0.264 0.321 0.098

I II III IV