Auction Formats, Bidding Behavior, and Price Outcomes
- sulla80
- Jul 10
- 15 min read
Updated: Jul 19
Auctions are the main place to buy rare collectibles such as coins and fine art, and auction design and behavioral factors can impact final prices. Today we will deviate from the usual topics of history and specific coins to focus on the mechanics of how we buy coins. What behaviors and psychological factors influence bidding and prices? What strategies are relevant for buyers?
A Note on Auction History
Auctions have been around for a while - Herodotus provides an early example, 6th century BCE Babylonia, of a system for auctioning of maidens for marriage:
"When the maidens grew to the age for marriage, they gathered these all together and brought them in a body to one place, and round them stood a company of men: and the crier caused each one severally to stand up, and proceeded to sell them, first the most comely of all, and afterwards, when she had been sold and had fetched a large sum of money, he would put up another who was the most comely after her: and they were sold for marriage. Now all the wealthy men of the Babylonians who were ready to marry vied with one another in bidding for the most beautiful maidens; those however of the common sort who were ready to marry did not require a fine form, but they would accept money together with less comely maidens."
-Herodotus, Histories 1.196Andocides, writing c. 440 – c. 370 BCE, describes the auctioning of tax‑farming contracts where the highest bidder won the right to collect taxes:
"You know what they are like; it is my belief that they meet there for a double purpose: to be paid for not raising the bidding, and to take shares in taxes which have been knocked down cheap."
Andocides, The Mysteries, 1.73.4For a Roman Republican example: Polybius 6.17.1-4 describes public sales in his long exposition on the Roman constitution; chapter 6.17 explains how censors leased state assets. In Republican practice each lease was cried out by a herald, bids were taken openly, and the contract went to the top offer.
The tools used in modern auction platforms have been worked on for thousands of years to sell valuable items.
Auction Formats and Bidding Options
Modern online auctions for unique, high-value items (like ancient coins and fine art) use a variety of features. This section provides an overview of common elements of auction platforms.
Proxy Bidding
In a (second-price) auction (e.g. eBay style, Leu style, biddr style), you use a proxy bid to submit a maximum bid and the system automatically bids the minimum necessary to lead, up to that max. In theory (game theory), this format has a dominant strategy: bid your true willingness-to-pay and let the proxy agent act for you. In practice, however, not all bidders trust or understand proxy agents, preferring the control of attending the auction, while others avoid proxy bidding out of fear that revealing their maximum too early invites shill bids or encourages others to drive up the price. Here's what proxy bidding looks like on biddr.
Proxy bidding generally increases the speed of the auction and can protect bidders from overpaying, and it might slightly reduce final prices compared to live bidding only (open out-cry). It takes discipline to place a bid and not watch an auction feverishly, but doing so can protect one from emotional overbidding. If everyone bids their true max via proxy, the price stops just above the second-highest valuation. In a live competitive auction, sometimes emotion or competition drives the top two beyond their initial valuations ("auction fever"), which can sometimes raise the winning price to extraordinary height. Buyers should be cautious in live formats - auction fever may push you past your valuation. Interestingly, empirical work classifying bidder types finds that those who use proxy agents (and do not keep adjusting their bids) often achieve the highest buyer surplus (savings between willingness to pay and price paid) on average (Cui et al 2008).
Pre-bidding (Absentee Bids): Many ancient coin auctions allow “pre-bids” or absentee bids before a live closing phase. Bidders can place early maximum bids days or hours in advance. Pre-bidding gives busy bidders a chance to participate early, but it also reveals interest in the item. A strong absentee bid can signal to others that the item is in demand, potentially deterring some competition or, conversely, attracting challengers if the current price is high. Some experienced bidders therefore place low absentee bids as a placeholder, or none at all, preferring to participate live where they can adjust in real time or via proxy bids when that option is available. Here's an example of the pre-bid from biddr.com:
Both pre-bidding and proxy bidding require trust in the auction house and their use of your price ceiling. Platforms AUEX, biddr, and Redpoint (live-bidder) all make it clear that auction bids are shared with the auction house for execution. So the auction house could mis-use the information and raise the price to meet or almost meet you maximum (shill bidding).
Hard Close vs Soft Close
The ending rule is critical in online auctions. A hard close auction ends at a predetermined time - when time runs out, the highest bid wins (e.g. most eBay auctions).
Research shows that soft-close formats generally drives higher selling prices than hard-close formats. This means that buyers in soft-close auctions are likely to pay a higher price. A study using a natural experiment on Yahoo auctions found that identical items sold with a soft-close rule earned 13–20% higher prices on average than those with hard deadlines (Glover et al 2012). The likely reason is that hard-close auctions make sniping a potentially valuable option. This means that some willing buyers don’t get a chance to counter-bid, keeping the final price lower than it could have been. (Glover et al 2012). This advantage comes when only bidders are holding back their best bid. This can also backfire if multiple snipers use extra-large bids as a strategy to ensure a win in the sniping - relying on all other bidders to be rational in their valuation of the item.
This is what that hard-close format looks like for a Leu auction (Ebay, Nomos and others are similar) with the clock running down:
A soft close (or dynamic closing) auction automatically extends the closing time if a last-minute bid arrives, allowing bidding to continue until no new bids come in. This is what the "soft close" looks like on biddr.com:
Bidder incentives and behaviors differ in these two formats: with a "hard close", bidders often engage in last-second bidding (sniping), whereas soft-close auctions eliminate the advantage of bidding at the last second. Platforms that use soft close see far fewer last-second bids, since any late bid gives others extra time to respond.
From the buyer’s perspective, hard-close auctions can occasionally be bargains - a savvy last-second bidder might win an item for less because a competing buyer’s ran out of time and other buyers failed to submit their highest bid.
Soft close rules lower the time advantage of a last-second bid, thereby encouraging open competition until the highest bidder truly wins. Consistent with this, multiple studies (including both field data and lab experiments) confirm that prices are higher when late bidding is deterred or when auctions stay open longer. (Gray et al 2009)
Buy or Bid sales
A popular variant for ancient coins on the "hard close" in which the seller publishes a wishful “buy” price typically a little above the dealer’s retail expectations. A collector who is willing to pay that figure can secure the item immediately. If you think the buy price is rich, you may leave a lower bid. When the sale closes, the highest bid that did not reach the buy price wins, usually at one increment above the second‑highest bid. The posted buy price doubles as a reserve; once someone pays it, the lot is removed from further bidding. Harlan J. Berk offers a buy or bid sale as an example.
"Fomo" or "fear of missing out" encourages bidders to end the auction at a price above what they might pay in other formats. The upside for buyers: this format does limit the risk of auction frenzy - uncontrolled bidding that could raise the price irrationally. It also means you have to rely on the price being set well by the auction house - this requires a knowledgeable team at the auction house and the buyer should ideally do their homework before bidding.
Auctions are Designed for Higher Prices
Buyers should be aware that professionally staged sales are designed to build trust and push prices higher - do your own research before assuming quality.
Presentation & descriptions matters: Also worth noting that auction houses vary in the quality of their photos and reliability of attributions. This can work for and against the buyer - an informed collector may recognize a mis-attributed coin or a coin that is not photographed to its advantage. Conversely without experience of the vendor one can overpay for a coin that may not be well represented by either the description or the photo. A coin in hand is not the same as a photo - reading weight, diameter, and paying close attention to the details of the description (tooled, smoothed, graffiti...) is essential to fairly value a coin. Seller reputation systems and information disclosure (like providing expert coin grading or provenance) also affect bidder confidence and willingness-to-pay, as documented in various studies. (Bajari et al. 2004) Venues that don't do well at presentation attract fewer bidders for an item and therefore the potential for a lower price.
Low starting bids are designed to attract competition, which may ultimately push the price higher. Low starting price and reserve prices attract more bidders (due to low entry cost), which can lead to higher final prices, whereas high reserves or starting bids can discourage bidders and lead to the item not selling (Cui et al 2008).
Ancient coin auctions get the best of all formats with modern platforms: pre-bid, proxy-bid, and live - leaving plenty of room for a few motivated buyers to drive the price to unprecedented levels in pre-bidding and bidding wars. The experienced buyer will just avoid engaging.
NOTE: if bidders were perfectly rational and bid their true maximum early (as proxy bidding allows) variations in format wouldn't change the outcome. However, behavioral realities diverge from this ideal.
Bidder Behavior: Sniping, Incremental Bidding, and Other Tactics
Online bidders exhibit a range of strategic behaviors that can work for or against their own interests. A comprehensive survey by Bajari and Hortaçsu (2004) noted that “frequently observed sniping, or last-second bidding behavior” is a major phenomenon in internet auctions.
Last-Second Bidding (Sniping): In hard-close auctions, a large share of bids often arrives in the final moments. On eBay, it’s routine to see significant bidding activity in the last minute or even last few seconds of an auction. Roth and Ockenfels (2002) famously compared eBay (hard close) to Amazon’s auctions (which at that time had a 10-minute extension soft close) and found stark differences: “Many bidders snipe on eBay… Late bids of this sort are much less frequent in auctions that are automatically extended (Amazon)… There is more sniping on eBay than on Amazon, and this difference grows with experience”. In other words, veteran eBay bidders are even more likely to bid at the last second than novices, indicating they’ve learned the strategic value of sniping. (Ockenfels et al 2006)
Does sniping pay off? Empirical evidence suggests it gives a small advantage to the sniper, but not a large one. Field experiments by Gray and Reiley in which identical items were auctioned with an early-bid strategy in one case and a snipe in another found at best a 2–3% lower final price for the sniper on average, however the difference was not always statistically significant. (Gray et al 2009) Another study found sniping had a very small effect when many participants in the market are experienced. This is certainly supported by personal experience at Leu and Nomos which attract a large number of experienced bidders and seem to often have high hammer prices using a hard-close auction format.
Incremental Bidding (Bid “Nibbling”): This is the strategy of placing successive small bids, essentially bidding step-by-step up to your limit rather than proxy bidding your maximum outright. Many less experienced bidders do this either out of caution or the thrill of competition. For example, a bidder might bid $100, then see they are outbid by an automatic proxy, then bid $200, outbid again, then $300, and so on. Incremental bidding can backfire – it telegraphs information about your interest and allows prices to ratchet up (Cui et al 2008). Research by Ockenfels and Roth noted that the presence of incremental bidders invites sniping by others as a counter-strategy. Essentially, if others see you gradually raising your bid, a smart rival will wait and strike once at a high amount to beat you. Incremental bidding also increases the risk of emotional bidding wars, where bidders exceed their value due to the “heat of the moment.” Despite its pitfalls, this behavior is common; Borle et al. (2006) even consider it a “pedestrian” bidding strategy of strategic importance in online auctions.
Some bidders use incremental bidding to discover the current top price (probing the proxy system to see at what point they become the high bidder), but this can trigger the proxy agent of a rival to immediately outbid them, driving the price up with no benefit to the incremental bidder.
Jump Bidding: In contrast to nibbling, jump bidders intentionally bid much higher than the current price increment – essentially placing a high bid early to signal strength. For instance, if an ancient coin is at $500, a jump bidder might suddenly bid $800 or $1000, even though no one has pushed it that high yet. This can be a strategic move: it discourages some competitors by making the next required bid much higher, and signals that “this bidder is serious.” Studies have found that experienced bidders sometimes employ jump bidding at early stages. It can shorten an auction and avoid prolonged competition, potentially saving time and hassle. However, jump bidding also reveals your willingness to pay (at least a lower bound of it) and might attract attention to an item.
In high-end art auctions, jump bids are often used by confident bidders to try to knock out the competition quickly. The effect on final price is ambiguous - a bold jump can either scare off weaker hands (leading to a lower price if no one else responds) or entice a showdown with another strong bidder (potentially driving even higher prices). Laboratory experiments suggest that jump bidding can increase the price when bidders are impatient (they “telegraph” their strength to try ending the auction sooner). In charity auctions, encouraging jump bids while discouraging sniping was shown to increase prices, making silent auctions perform nearly as well as live English auctions.
Bidder Herding and Other Behaviors: In high-end auctions, reputation and signals can influence behavior. For example, if an ancient coin auction shows that a respected dealer or a bidder with a strong track record has placed a bid (some platforms display bidder aliases or ratings), others might infer the coin is especially valuable. This can trigger herding behavior (everyone wants what an expert wants) or, conversely, intimidate some from bidding further. There’s also the “winner’s curse” phenomenon in auctions of items with uncertain value - essentially, the worry that the winner might be the one who overestimated the item’s true value the most. Empirical studies confirm the winner’s curse exists in online auctions for unique items. In coin and art auctions, the winner’s curse can be relevant, especially for newcomers who haven’t calibrated their valuations. Experienced bidders may bid more conservatively as a result.
Experience is a key factor in bidding behavior: a study of internet art auctions found that with experience, bidders significantly lower their bids, by about 26% after ten or more auctions on average (Pownall et al 2013). This suggests that new bidders tend to overbid or get caught in auctions fever, whereas seasoned bidders learn to bid more cautiously and avoid overpaying. They might learn to snipe instead of bidding early, or learn to set a firm maximum and not exceed it. The learning effect here is important for high-end markets - bidders who repeatedly participate gain knowledge of true market values and proper strategies, leading them to better outcomes over time.
Currency & Shipping considerations in pricing: it is easy for bidders in an auction to ignore the many costs that might be associated - beyond the hammer price. Buyers fees, shipping, currency exchange, import/expert fees, and other costs have been high on the minds of auction goers and auction houses with international tariff policy in the US changing and relative values of currencies being more volatile over the last few months. This chart tells the story nicely for those buying with US dollars in auctions in other parts of the world (Source: https://www.google.com/finance/quote/USD-EUR?window=YTD). The USD has performed similarly vs a large basket of currencies.
There is no shortage of "money illusion research" to show that the nominal (face) value of money affects consumer perceptions of its real value i.e. 1000 Yen is more expensive that 10 US dollars. Non-obvious fees, and changing exchange rates only add to this mis-perception of cost. Some platforms are better than others in showing exchange rates and fees. Understanding the fully loaded cost is important in setting your bid limits to avoid over-paying.
Under-represented auctions and/or over-represented coins: In contexts where the likely competition is high, an auction can fetch a higher price; where demand is weak or supply strong buyers can find bargains. A savvy collector can find opportunities in several ways e.g.
when multiple of the same coin show up and the focus is on the better one or two in the auction
below average coins at a normally high priced auctions,
a good nice set of coins in an auction that is competing with a higher profile auction that pulls away buyers
an auction that is not well publicized
an auction scheduled at a time when some buyers are not able or interested in participating (national holidays that differ across regions)
a niche coin in a general auction that escapes the specialist buyers notice
Some of these factors explain why we often see auction houses buying selling each other's coins. It also means that collectors should be thoughtful about where to sell their coins in order to maximize the value of their collection. One venue may not be appropriate for your whole collection.
Conclusions
Know the coin & research your price: There is no substitute for knowing the coin that you are buying. In the niche and wide ranging field of ancient coins, bid prices are often influenced by buyers who do not know what they are buying. You cannot rely only other bidders to decide if a price is fair or not. For many of the more popular coin types - another one will come along soon.
After a deep dive on auction mechanics and bidder behaviors the main conclusion can be summed up as this set of recommendations.
Auction sophistication: Auctions have been around for thousands of year - modern platforms are well designed to get to maximum participation, engagement and encourage top sales prices. Buyers should understand that platforms are intentionally structured to increase bidding intensity - bidding strategy matters.
Do your homework: know the coin, know the auction house, know the hidden fees and rules of bidding, know the currency exchange & import fees.
Avoid emotional over-bidding ("auction fever") set a price and stick to it
Hold your cards close: Revealing your interest in a coin changes the price so hiding your interest as much as possible makes sense:
If you do pre-bid don't do it incrementally - incremental bidding can work against the bidders interest by increasing the risk of bidding wars and encouraging use of sniping.
Use a proxy bid (with a trustworthy auction house) as the best bet for an auction that uses a soft-close approach (most ancient coin auctions).
Sniping can be useful in hard-close auctions: A last minute bid or "snipe" doesn't benefit you in an environment where many are sniping and/or many have put in a full value pre-bid or proxy bid. It is a risky strategy to use oversized snipe bids - it will work against you if there are others using this strategy to lock in a win. Snipes may have a modest positive impact in auctions with few participants and less experienced bidders.
The biggest opportunities for buyer's price advantage exist in sparsely attended auctions, less expert sellers, and coins that don't attract the right buyers.
incorrectly attributed coins
coins with low quality photographs
coins in the wrong place in the listing catalog
overlooked coins in a field of similar coins
smaller auctions that compete at the same time as a high profile auction
specialist coins at a non-specialist venue
References:
Kellgren, Martin. 2023. “The Impact of Soft‑Close Auctions on Bid Behavior: A Randomised Controlled Experiment.” Master’s thesis, Graduate School, School of Business, Economics and Law, University of Gothenburg.
Chen, Yan, Peter Cramton, John A. List, and Axel Ockenfels. 2020. Market Design, Human Behavior, and Management. NBER Working Paper 26873. Cambridge, MA: National Bureau of Economic Research.
PON Staff. 2019. “Negotiation Research You Can Use: Inoculate Yourself against Auction Fever.” Program on Negotiation at Harvard Law School, January 31.
Adam, Marc T. P., Gillian Ku, and Ewa Lux. 2018. “Auction Fever: The Unrecognized Effects of Incidental Arousal.” Journal of Experimental Social Psychology 77: 186–198.
Pownall, Rachel A. J., and Leonard Wolk. 2013. “Bidding Behavior and Experience in Internet Auctions.” European Economic Review 61: 14–27.
Glover, Brent, and Yaron Raviv. 2012. “Revenue Non‑Equivalence between Auctions with Soft and Hard Closing Mechanisms: New Evidence from Yahoo!” Journal of Economic Behavior & Organization 81 (1): 129–136.
Malaga, Ross A., Dawn Porter, Keith Ord, and Bonnie Montano. 2010. “A New End‑of‑Auction Model for Curbing Sniping.” Journal of the Operational Research Society 61 (8): 1265–1272.
Gray, Sean, and David H. Reiley. 2009. “Measuring the Benefits to Sniping on eBay: Evidence from a Field Experiment.” Working paper, New York University and University of Arizona.
Collector’s Universe Forum. 2009. “Auction Houses and 15 Minute Rule.” Discussion thread, Collector’s Universe (PCGS) Message Boards.
Cui, Xiling, Vincent S. Lai, and Connie K. W. Liu. 2008. “Consumer Behavior in Online Auctions: A Critical Review.” Electronic Markets 18 (4): 344–361.
Wertenbroch, Klaus, Dilip Soman, and Amitava Chattopadhyay. 2007. “On the Perceived Value of Money: The Reference Dependence of Currency Numerosity Effects.” Journal of Consumer Research 34 (1): 1–10. .
Borle, Sharad, Peter Boatwright, and Joseph B. Kadane. 2006. “The Timing of Bid Placement and Extent of Multiple Bidding: An Empirical Investigation Using eBay Online Auctions.” Statistical Science 21 (2): 194–205.
Ockenfels, Axel, and Alvin E. Roth. 2006. “Late and Multiple Bidding in Second‑Price Internet Auctions: Theory and Evidence Concerning Different Rules for Ending an Auction.” Games and Economic Behavior 55 (2): 297–320.
Ariely, Dan, Axel Ockenfels, and Alvin E. Roth. 2005. “An Experimental Analysis of Ending Rules in Internet Auctions.” RAND Journal of Economics 36 (4): 890–907.
Ku, Gill, Deepak Malhotra, and J. Keith Murnighan. 2005. “Towards a Competitive Arousal Model: Auction Fever and the Winner’s Curse.” Organizational Behavior and Human Decision Processes 96 (2): 89–103.
Bajari, Patrick, and Ali Hortaçsu. 2004. “Economic Insights from Internet Auctions.” Journal of Economic Literature 42 (2): 457–486.
Roth, Alvin E., and Axel Ockenfels. 2002. “Last‑Minute Bidding and the Rules for Ending Second‑Price Auctions: Evidence from eBay and Amazon.” American Economic Review 92 (4): 1093–1103.
Soman, Dilip, Klaus Wertenbroch, and Amitava Chattopadhyay. 2002. “Currency Numerosity Effects on the Perceived Value of Transactions.” INSEAD Working Paper.
Budish, Eric, and Leslie Takeyama. 2001. “Buy Prices in Online Auctions: Irrationality on the Internet?” Economics Letters 72 (3): 325–333.
Comments