The Inconvenient Truth About Secondary Markets, Part II.

By Massimo Franceschet and Sparrow Read

Visual conversation at dada.art

Watch the Working Group discussion: The Inconvenient Truth About Secondary Markets part I.

Introduction

For the past 6 months, members of the DADA community have been exploring the relationship between rare digital art and marketplaces on the blockchain.

This work is part of the Invisible Economy: a vision for radically separating art from the marketplace so that artists are able to focus on making art, and satisfying their intrinsic needs without the distraction of needing to market or the pressure to sell their art.

For us, DADA is a place where people make art together without expecting remuneration, motivated not by extrinsic rewards like money or status but by intrinsic rewards such as the joy of making art, and a sense of autonomy, validation, self-development, belonging, and a higher purpose. By allowing artists to focus on their intrinsic desire to make art, a sustainable economy can be created based on interdependence rather than competition.

Background

The blockchain art market has been steadily growing while a handful of art marketplaces founded in 2018 have become well known in the NFT space. That year, a small group of early adopter artists exploring this technology partnered with some of these tech start-ups to tokenize their artworks using smart contracts on the Ethereum blockchain.

At the beginning of 2020, some of those early adopters in the blockchain art community celebrated a victory for artists and art by implementing a secondary sales royalty, enforced by smart contracts, across all of the major platforms. The problem of artists’ royalties (or artist resale rights, as they are often known) has been around for well over 100 years in the traditional art market; and has been a persistent problem that the art world as a whole has struggled to solve. With the advent of distributed ledger (blockchain) technology, implementing secondary sale royalties for artists was now attainable and all of the major rare art marketplaces agreed to a 10% artist royalty on secondary sales.

The secondary markets in the blockchain art space began gaining momentum approximately fourteen months ago, in the second half of 2019, and the trend has been towards increasingly high returns on investment sales on the secondary market. These have often made headlines and have, arguably, attracted more speculative investors.

Throughout 2020 we have seen this secondary market grow significantly. In the context of the Invisible Economy, which is separating art from the marketplace, we sought data on what the current system did for artists. In addition to understanding the current situation, we also wanted to try to peer into the future to see where the trend would lead to for artists in general.

It has generally been thought that developing secondary markets — particularly with the 10% royalties — is a ‘good thing’ for artists. But what does the data tell us?

Our Thesis

By nurturing a speculative secondary market based on investment and return on investment, in which artists are transferring 90% of their source of wealth to collectors in perpetuity, it is all but inevitable that at some point, collectors as a group will create more wealth, and hold more power than artists as a group.

These were the questions we posed to the data. We were interested to see how much time it would take to reach an inflection point where this cross-over would occur. The following shows both the current situation, as well as a predictive model to attempt to gauge how quickly or slowly we will reach this cross-over point.

Data Analysis

In this data analysis we assume a crypto art marketplace with the following price scheme:

• primary market (sales from the original artist to the first collector): 15% of the fee goes to the gallery and 85% to the artist.
• secondary market (sales from collector to collector): 10% of royalties to the original artist and 90% to the collector.
• moreover, there is a simple 3% transaction fee for all primary and secondary purchases, paid by the buyer (collector).

We define:

• the volume earned by artists (artist volume) as the net prices (without fees) of sales on the primary market plus the royalties received by artists on the secondary market;
• the volume earned by collectors (collector volume) as the net prices (without royalties) of the sales on the secondary market;
• the overall volume generated by the gallery (gallery volume) as the sum of volumes generated by artists and collectors.

All prices are defined in USD with the ETH/USD exchange rate on the day of the sale.

We pose the following data queries:

• is the market concentrated or equidistributed?
• how did the market concentration evolve over time?
• what is the share of artist and collector volumes as of today?
• how did the artist and collector volumes change over time?
• when in the future are the artist and collector volumes going to break even?
• what if the royalty percentage for artists would be larger?

We used the SuperRare dataset from day one (April 2018) until today (November 2020) to try to answer these queries.

Concentration of Volume Among Sellers and Buyers

We will use two well-known econometrics tools to investigate wealth concentration: The Lorenz curve and the Gini index.

The Lorenz curve plots the cumulative percentages of total income received against the cumulative percentage of number of recipients, starting with the richest individual. The Gini index measures the area between the Lorenz curve and a hypothetical line of absolute equality, expressed as a percentage of the maximum area under the line. Thus a Gini index of 0 represents perfect equality, while an index of 100 implies perfect inequality. Gini index is typically adopted to measure the extent to which the distribution of income among individuals within an economy deviates from a perfectly equal distribution (the country ranking by Gini index on income).

Here we applied the Gini index to the share of sales volume received by sellers or spent by buyers. It turns out that the crypto market is extremely concentrated among few artists and even fewer collectors: if crypto art were a country and sales the income of people, it would be much more concentrated than any other country at any time in history! The market was not that concentrated at the beginning of crypto art gallery life, in particular for collectors. Collector concentration overtook artist concentration in December 2018 and since then it was always significantly higher.

Artist and Collector Volumes

“The artists are the ones who make the magic happen, so be respectful to all SuperRare artists even if you are not a fan of their work”. SuperRare Community Guidelines

Of the total gallery volume on SuperRare (primary and secondary sales):

• 78% goes to artists (fee-free primary sales plus royalties on secondary sales)
• 22% goes to collectors (royalty-free secondary sales)
• 80% of the gallery fees are paid by artists (with the 15% fee of primary market sales) and 20% are paid by collectors (with the 3% transaction fee)

Moreover:

• of all sales, 85% are primary sales and only 15% are secondary sales (the secondary market is still underdeveloped)
• out of every 100 artworks bought by collectors only 17 of them are resold (most collectors have strong hands)
• collectors resell at 1.68 times the price they buy (some collectors are art speculators and buy low to resell high)

Hence, we observe an artist dominance where the volume earned by artists is far bigger than that gained by collectors. Over time, the artist dominance was quite stable and above 90% until October 2019, when it started to decline in favor of a rise of the collector share. Assuming that this trend continues, when do the shares earned by artists and collectors break even?

Predicting Artists’ Dominance

Let’s fit a linear regression model. The linear regression model has time (number of months) as an independent variable and the share of volume earned by collectors as a response (dependent) variable. It predicts the following linear model with good quality of fit (R-Squared is 0.88):

collector share = 0.095697 + 0.012497 ⋅ month

This means that according to the fitted model, every month the collector volume increases by 1.25%. Hence we can predict that the collectors’ volume share will be 0.5 and therefore equal to the artists’ volume share in about 32 months from October 2019, or in 19 months (1 year and 7 months) from now, that is, by June 2022. Unless the artist royalty share is increased soon.

What If Scenarios

When will the break-even happen between artist and collector volumes if we increase the royalty percentage? Let’s check:

• assuming an artist royalty of 10%, the slope of the collector regression line becomes 1.25% and the break-even with artists will be 19 months from now.
• assuming an artist royalty of 20%, the slope of the collector regression line becomes 1.11% and the break-even with artists will be 24 months from now.
• assuming an artist royalty of 50%, the slope of the collector regression line becomes 0.69% and the break-even with artists will be 64 months (over 5 years) from now.
• assuming an artist royalty of 90%, the slope of the collector regression line becomes 0.14% and the break-even with artists will be 352 months (29 years) from now.

Conclusion

One of the reasons we posed these questions and wanted to see what the data would tell us is to get a sense of whether the crypto art market is achieving some of the many aims that are often stated as ideals in the space: empowering artists, the democratization of art, and being something truly different from the old, broken art market.

As they say, “the road to hell is paved with good intentions.” It is important to monitor the outcomes we produce and check whether we are achieving what we set out to achieve. This has been a lesson we learned in the Token Engineering Academy. If we don’t do this, we risk following the same path as the centralized networked platforms where the trend has been “attract then extract, cooperate then compete” seemingly regardless of the original aims of those platforms.

The crypto art space appears to be on an accelerated path to replicating the problems we see in the traditional art market, such as:

• from 2 million art school graduates, only 200,000 remain working artists;
• the fact that at least three-quarters of working artists (even in ‘rich’ countries like the UK and US) cannot live from their work alone;
• a concentration of sales among a very few ‘top’ artists (in the traditional art market currently 64% of sales come from the top 3 artists);
• driving an overall lack of diversity in the art world (87% of artists represented in museums are men and 85% of them white).

We believe that this state of affairs is not inevitable. It is an outcome of the decisions we make in creating the systems in which art operates.

Our theory is that this situation arises because we are using only one metric of “success” and focus solely on sales. This could be a symptom of crypto culture and its focus on price, all-time-highs (ATH), and return-on-investment (ROI). But this leads to an effect often referred to as the Pareto Principle or power-law distribution, named after the Italian economist, Vilfredo Pareto, who in 1895 noticed that about 80% of Italy’s land belonged to 20% of the country’s population.

We think that we can do better and that we can build systems that achieve the outcomes we say we want. We think that we are still really early. Early enough to change how this plays out. But we can’t just hope that things will be different simply because we are using the latest cutting edge technology. We are seeing the same systems being re-created, so inevitably they will produce the same results.

We have spent the past few months talking about ways to ensure different results. Even without adopting the Invisible Economy as a whole, there are many things that can be done. These may include looking at incentives and aligning them not just with making money but other things of value. Platforms could rethink the allocation of rewards and status, the role of curation, and how works are surfaced. Understanding that the systems that we build encourage specific behaviors can help us build and tweak systems for fairer, more equitable, and more community-driven outcomes, better aligned with the ideals we all hold.

We are deeply grateful to SuperRare for making the dataset used in this analysis available. They have been open, helpful, and willing to explore these topics with us.

Massimo Franceschet is an artist and data scientist.

Sparrow Read is an artist and technologist.