First Offers on eBay Motors

Document Sample
First Offers on eBay Motors Powered By Docstoc
					                      First Offers on eBay Motors

Jong-Rong Chena, Ching-I Huangb,∗, Chiu-Yu Leea
 Institute of Industrial Economics, National Central University, Jhongli City 32001,
    Department of Economics, National Taiwan University, Taipei City 10055, Taiwan


We empirically investigate the factors which influence a buyer’s behavior in making
an offer to purchase a car listed on eBay Motors using the “Best Offer” option. Our
results indicate that a buyer’s first offer is affected by observed information, including
the number of buyers making an offer for the same item and the length of time since
the start of the listing.

Key words: eBay, best offer
JEL Classification: L81

    Corresponding author. Tel.: +886 2 23516941; fax: +886 2 23511826.
    E-mail address: (C.-I. Huang)
1. Introduction

     Although there have been several studies investigating the critical factors that
affect consumer behavior in online auction websites (e.g., Ockenfels et al., 2006),
there is still a lack of those examining buyer behavior on “Best Offer” in theses
websites. To the best of our knowledge, this is the first empirical study on this topic,
where we focus on a buyer’s first offer to purchase a car on eBay Motors.

EBay added the “Best Offer” feature as an optional selling format in 2005. This
feature enables a buyer to negotiate with the seller for a price lower than the listed
“Buy It Now” (BIN) price. When a seller lists an item in the “fixed price format”,
he/she can choose to include the “Best Offer” feature at no additional cost. A buyer
can make up to ten offers per listing with this feature.1 The seller can accept the offer,
reject the offer, make a counteroffer, or simply ignore the offer. When the seller
accepts an offer or a buyer accepts a counteroffer, the transaction is completed.

There are two primary reasons for focusing on a buyer’s first offer during the listing
period. First, our data reveal that most of the accepted offers (87%) are actually the
buyer’s first offer. Second, a buyer’s second or later offers depend on the seller’s
responses to earlier offers. Nonetheless, our data do not allow us to know details of
these responses. By restricting our attention to first offers we can safely exclude these
later responses.

2. Data and methodology

      We collected data on all Toyota Camry cars listed on eBay Motors during the
period from June 18, 2008 to March 6, 2009. The original dataset included 4,347
listings. Of these, 650 used the Best Offer selling format. The “offer history’’ of these
650 listings gave us information about each buyer’s offers.

The value of a car varies with many observable characteristics, such as age, mileage,
and mechanical condition. The Kelly Blue Book (more commonly known as the “blue
book”) is a commonly used reference for estimating the value of a car. Because our
focus is the impact of trading system on offers, not the impact of observed
characteristics, the blue book price is used to control for heterogeneities. Specifically,
we normalize a price (seller’s listed Buy-It-Now price or buyer’s offer) by first
dividing the price by the blue book price for the listed car and then take a logarithm of

    The maximum number of offers for products other than cars is three.

the ratio. Therefore, the normalized price can be interpreted as being the percentage of
deviation from the blue book price.

We specify the full model as follows:


     ln(BIN/BLUE), CLEAR, WARRANTY, SELLER_SCORE,                            
                                                                                     (1)
     TIME, WEEKEND, MORNING                                                  
                                                                             

The variables in Eq. (1) and the expected sign are summarized in Table 1. The first six
variables are the characteristics of a listing, while the rest indicate the characteristics
of a buyer’s first offer within a listing. We have 978 observations of first offers among
the 1,493 offers in our data.

3. Empirical Results

      The model is estimated with ordinary least squares regression. The t-values are
computed based on robust standard errors, allowing for correlated error terms for
offers within the same listing. Table 2 displays results of six different regression
models on the determinants of offer prices. After deleting incomplete samples, we are
left with 754-784 observations2.

All the estimated coefficients have the expected sign. The results are robust across all
specifications we tried, including other setups not reported here. Our empirical results
suggest that buyers use the available information during a listing period to determine
the offer price. In particular, both the length of time since the start of the listing and
the number of other buyers having made an offer in the same listing have significant

On average, buyers begin to make offers roughly 144 hours after a car is listed3. The
time length since the start of the listing is significant and negatively related to the
offer price. With every passing hour, the price that a buyer is willing to offer is
reduced by 0.08%. The negative sign has two possible explanations. An individual
with a stronger desire to buy tends to make an offer earlier and the offer price is
usually higher. Alternatively, a buyer is likely to offer a lower price toward the end of

  Some sellers refuse to disclose the buyer’s offer in the offer history. We thus lose 86 observations.
Also, we could not obtain a blue book price for some cars, mainly due to unavailability of the owner’s
ZIP code or vehicle’s age, meaning we lose another 119 observations.
  Average listing duration in our sample is 11.39 days.

a listing period because the seller is willing to accept a relatively lower offer to avoid
no transaction.

The buyer can observe an item’s offer history during the listing period, which includes
the time, status and incomplete buyer ID for all previous offers made for this item.4
Consequently, a buyer can infer the demand from the number of buyers who have
made an offer for this item. We find that the number of buyers who have made an
offer on the same item has a significantly positive effect on a buyer’s offer price.
Similarly, the first buyer to make an offer after listing tends to offer a significantly
lower price. Both these tendencies indicate that the inferred demand affects a buyer’s
decision. However, since these two covariates are highly correlated, when both are
included in the regression, only one of them is significant.

It should also be noted that the seller’s reputation is recognized as a critical issue in
Internet auctions (Melnik and Alm, 2002; Livingston, 2005; Houser and Wooders,
2006). Our analysis of the offer prices shows similar results. The seller’s feedback
score on eBay increases offer prices.

The buyer’s eBay feedback score has a significantly positive effect on the offer price.
This finding is contrary to findings for Internet auctions. For instance, Houser and
Wooders (2006) show that buyer’s reputation has no effect on price. The score is not
only representative of a buyer’s reputation for doing business on eBay, but it can also
indicate a buyer’s experience on eBay. A buyer with a good reputation can expect the
seller to be willing to accept a relatively lower price since he is more likely to
honestly complete the transaction process. An experienced buyer can more accurately
evaluate the value of a car and will make a higher first offer, thereby to avoid
transaction costs due to being rejected and having to make another offer. Our
regression results suggest the latter effect dominates the former one.

Our results also show that offers made between 6 A.M. to 10 A.M. Pacific Time tend to
be significantly higher than those made at other times. Offers proposed on weekdays
are also higher, although the difference is insignificant. The reason could be that
buyers making an offer on weekday mornings seem to have stronger desire to buy a

Finally, a 1% increase in ln(BIN/BLUE) is associated with approximately a 1%
increase in ln(OFFERPRICE/BLUE), which indicates that, when a seller sets a higher

    The price and buyer’s compete ID are only available to the seller.

BIN price, a buyer’s offer price will be higher. Nonetheless, the endogeneity of the
BIN price may be a concern for establishing a causal relationship. When a car is of
better quality, which can be observed by the seller and the buyer, but unobserved by
econometricians, the seller will post a higher BIN price and the buyer offer a higher
price. Since our focus is on the trading system, not on the choice of the BIN price, the
latter is regarded as a control variable. We do not treat the endogenous problem.

4. Conclusion

      Our study contributes to the understanding of the buyer’s behavior using the Best
Offer selling format on eBay Motors. We find that the length of time the offer is made
after the start of the listing is significantly related to the offering price. Although the
value of other buyers’ offers is unknown, the number of buyers who have made offers
is indicative of demand and affects the potential buyer’s offer. In future research we
could extend our work to study the interactions between the buyer and the seller after
the first offer.


Houser, D., and J. Wooders, 2006, Reputation in Auctions: Theory, and Evidence from
   eBay, Journal of Economics and Management Strategy 15, 353-369.
Livingston, J. A., 2005, How Valuable Is a Good Reputation? A Sample Selection
   Model of Internet Auctions, Review of Economics and Statistics 87, 453-465.
Melnik, M. I., and J. Alm, 2002, Does a Seller's Ecommerce Reputation Matter?
   Evidence from Ebay Auctions, Journal of Industrial Economics 50, 337-349.
Ockenfels, A., D. Reiley, and A. Sadrieh, 2006, Online Auctions, in: Terrence J.
    Hendershott, eds., Handbook of Economics and Information Systems, Vol.1
    (Netherlands, Amsterdam) 571-628.

Table 1. Summary of statistics and expected sign
Variables                 Definition                                       Mean     Std. Dev.
Listing characteristics
BIN (in $)                Buy It Now price (listed price)                  11826.13 7963.13
BLUE (in $)               Kelly Blue Book price                            11202.54 6751.02
ln(BIN/BLUE)              Log of (BIN price/ Kelly Blue Book price)        0.09     0.33        +
CLEAR                     Dummy =1 if the title is clear                   0.78     0.41        +
WARRANTY                  Dummy = 1 if vehicle is still under warranty     0.31     0.46        +
SELLER_SCORE              Seller’s feedback score on eBay                  347.64   1056.03 +
DEALER                    Dummy = 1 if the seller is a dealer              0.50     0.50        ?
Offer characteristics
OFFERPRICE (in $)         Buyer’s first offer price                        8041.23 6584.61
ln(OFFERPRICE/BLUE) Log of (first offer/ Kelly Blue Book price)            -0.47    0.87
PRIOR                     Number of buyers having made an offer            1.92     2.33        +
                          Dummy = 1 if the offer is the first one of the
LISTING_FIRST                                                              0.33     0.47        -
BUYER_SCORE               Buyer’s feedback score on eBay                   73.64    200.13      ?
                          Length of time since the start of the listing
TIME                                                                       144.29   115.88      -
                           (in hours)
WEEKEND                   Dummy = 1 if the offer is made on weekend        0.30     0.46        -
                          Dummy = 1 if the offer is made between 6-10
MORNING                                                                    0.23     0.42        +

Table 2. Empirical result

                             Model 1        Model 2    Model 3    Model 4     Model 5      Full model
                             -0.50***       -0.55***   -0.58***   -0.58***    -0.69***     -0.63***
                             (-7.67)        (-7.77)    (-7.13)    (-5.01)     (-7.33)      (-5.67)
Listing characteristics
                             1.04***        1.12***    1.10***    1.05***     1.07***      1.07***
                             (7.13)         (12.46)    (12.14)    (10.38)     (10.14)      (10.13)
                                                                  0.07        0.06         0.06
                                                                  (0.81)      (0.62)       (0.67)
                                                                  0.14**         0.13**    0.14**
                                                                  (2.43)      (2.31)       (2.43)
                                            0.0002**   0.0002**   0.0002**    0.0002**     0.0002**
                                            ( 2.36)    (2.20)     (2.18)      (2.42)       (2.36)
                                            0.12**     0.13**     0.07        0.07         0.07
                                            ( 2.07)    (2.16)     (1.24)      (1.21)        (1.31)
Offer characteristics
                             0.02*          0.03*      0.03**                 0.04***      0.03**
                             (1.92)         ( 2.37)    (2.42)                 (2.57)       (2.24)
                             -0.11          -0.09      -0.09      -0.18***                 -0.11
                             (-1.36)        (-1.26)    (-1.22)       (2.17)                (-1.46)
                                                       0.0002**   0.0002**    0.0002**     0.0002**
                                                       (2.28)     (2.23)      (2.06)       (2.13)
                             -0.0004        -0.0008*   -0.0008*   -0.0007*    -0.0008*     -0.0008*
                             (-0.96)        (-1.71)    (-1.68)    (-1.68)     (-1.76)      (-1.84)
                                                       -0.04      -0.03       -0.04        -0.04
                                                       (-0.68)    (-0.59)     (-0.71)      (-0.64)
                                                       0.14**     0.13**      0.13**       0.13**
                                                       (2.17)     (2.02)      (2.13)       (2.07)
R-squared                    0.1599         0.2318     0.2377     0.2412      0.2424       0.2450
Number of offers             784            761        754        754         754          754
Note: T-values are given in parentheses. Statistical significance at the 1%, 5% and 10% levels were
denoted with ***, ** and *, respectively.


Shared By: