Does spending decrease as purchasing power declines?

ecommerceLast year Peter Coy illustrated what a deflationary economy looks like (such as the Bitcoin economy) and explained how this impacts consumer spending (and lending).

Depending on what peak someone may have bought at, the very reverse happened this year, with prices denominated in bitcoin rising by perhaps as much as 65% (a full analysis should probably also adjust a couple percent to include CPI).

Though, to my knowledge there are no products actually denominated in bitcoin (yet).

So then, did spending habits change over the course of the year?

Not really, users as a whole still preferred to simply hold onto coins either because they had low time preferences with future expectations of large price appreciation and/or they were ‘underwater’ in coin (e.g., they bought at a peak).  Off-chain transactions on Coinbase did not see much of a difference (yet) either.

Economic theory suggests that consumers prefer a medium of exchange with stable purchasing power and in practice that seems to be the case.

For instance, on January 9, 2014 online retailer Overstock.com began accepting bitcoin as payment.  In the first two months it generated $1 million in bitcoin payments and through May the tally had grown to $1.6 million in bitcoin payments.

According to a new story yesterday, Overstock announced it would likely generate $3 million in bitcoin payments this year (though they do not specify how many bitcoins that is altogether).  This is in contrast to the estimates at the beginning of the year:

The figures are notable given that the e-commerce company had issued a wide range of potential estimates for its first-year bitcoin sales over the course of the year. In March, CEO Patrick Byrne suggested Overstock was on pace to achieve $10m–$15m, or even $20m, in bitcoin sales.

Such estimates were also below the $5m Byrne said Overstock originally anticipated, though on par with those suggested by Overstock chairman of the board Jonathan Johnson in interviews.

Altogether approximately 11,100 customers paid with bitcoin this past year at Overstock — these customers spent an average of $273 in bitcoins.  That means that after the initial power law from the first couple months, roughly $200,000 in bitcoin sales occurred from March onwards, or roughly $6,700 per day.

If bitcoin denominated prices had stayed the same, would that have increased the amount purchased?  Perhaps, but as articulated by both Robert Sams and Yanis Varoufakis, bitcoin stability is perpetually ephemeral and perhaps the only solution is to switch the monocoin ledger and adopt a dual currency ledger design instead (a topic for another day).

Besides a decline in purchasing power, is there anything else that may have caused this?

In chapter 11, pages 181-182 I explored another reason (see this image): demographics.  Most (60%) of the customer base of Overstock are female and as we know empirically, there are very few females that inhabit the Bitcoin ecosystem.  Perhaps this will change in time, so what are other datum in this exhibit?

Specifically, what does Overstock do with these coins?  One redditor looked through the most recent 10-Q filing and found:

At present we do not accept bitcoin payments directly, but use a third party vendor to accept bitcoin payments on our behalf. That third party vendor then immediately converts the bitcoin payments into U.S. dollars so that we receive payment for the product sold at the sales price in U.S. dollars.

[…]

We have also begun accumulating bitcoin in an amount of approximately 10% of the amount of our bitcoin-denominated sales as well as other cryptocurrency.

[…]

We hold cryptocurrency denominated assets such as bitcoin. We currently consider these holdings to be investments and include them with other long-term assets in our Consolidated Balance Sheets. Cryptocurrency denominated assets were $346,000 and zero at September 30, 2014 and December 31, 2013, respectively … Losses on cryptocurrency holdings were $50,000 during the three and nine months ended September 30, 2014. There were no losses on cryptocurrency holdings for the three and nine months ended September 30, 2013.

Or in other words, Overstock.com sells all but 90% of the coins it receives and puts the remaining portion onto its books as an investment, which saw a loss of $50,000 through Q3.  Perhaps this reverses next year if there is another run up in prices.

In addition, the coin sales created (a marginal) sell side pressure on the market through the intermediary, Bitpay, the largest payment processor in this space.

What changes did Bitpay see this year?  In a recent profile they noted that:

BitPay, the largest and oldest bitcoin payment processor with a daily volume of $1 million bitcoin transactions supporting more than 44,000 merchants, stated in an email exchange to CCN that more than 4,400 of their merchants keep all of their settlement in bitcoin, almost 18,000 keep some of their settlement in bitcoin while the remaining 22,000 convert it all to fiat.

While the amount of merchants accepting bitcoin more than quadrupled this year, the amount of retail commercial transactions did not.  Because Bitpay re-uses the addresses for purchases, it is possible to monitor them for inflows.  And over the past 6 months, there has not been a significant change: roughly 2,000 bitcoins in aggregate (+/- 200) are received by Bitpay each day.  In fact, they have been receiving approximately the same $1 million in bitcoin transactions since May. [Note: at current market prices, even 2,200 bitcoins does not equal to $1 million thus a contradiction, which can only be cleared if/when Bitpay releases its methodology]

Because of the ecosystem still lacks a ‘circular flow of income,’ in return Bitpay sells these coins to other inventory providers such as financial institutions, family offices and exchanges (detailed here).  This further creates sell side price pressure and if there is not a corresponding increase in speculative or transactional demand in bitcoins, effectively lowers the purchasing power of a coin.

For instance, last Wednesday, December 10th, Microsoft announced that it had added bitcoin as a payment vehicle for games and apps.  The price rallied 10% in the course of an hour yet subsequently declined to pre-rally prices.  Why?

As analyst Raffael Danielli explained to me, on the one hand, Microsoft under the new CEO — Satya Nadella — seems to push deliberately into areas at the forefront of the tech sector.  Accepting bitcoin is an item on their list that can easily be implemented and subsequently crossed off (e.g., a cheap point in terms of risk / reward due to the usage of an intermediary).

On the other hand, if people are less willing to spend Bitcoin while ‘underwater’ this can lead to more ‘bad’ news regarding a lack of consumer adoption.  For example, one could see a correlation between Xbox One’s less-than-stellar sales and losses against the Playstation 4 (PS4 is outselling 2:1), versus the need to get some kind of PR spark before the Christmas shopping spree.  Similarly Time magazine’s announcement today probably will only produce a temporary marginal increase in bitcoin activity and was likely done with similar motivations (positive PR before holidays) because Time been hit hardest (it’s 2012 sales of single-issue copies declined 27%, the most across the entire industry and it laid off 5% of its workforce in early 2013).

Yet most bitcoin holders are probably not the usual demographic of paper magazine subscribers.  Or as one droll redditor explained:

Venn diagram of people who use bitcoin and people who subscribe to print magazines: OO

Perhaps market participants as a whole see this too or perhaps they recognize that even if there was an upsurge in bitcoin usage to Microsoft product lines (which we can monitor as Microsoft is using Bitpay), those coins will ultimately put sell side pressure because there is no circular flow of income.  And again, without a corresponding amount of speculative or transactional demand, the price of a bitcoin could decline as would its purchasing power.

There is never a dull moment in this space, perhaps 2015 will create new patterns to analyze.

 

[Special thanks to Jop Hartog and Jonathan Levin for their feedback and information this past month]

Leave a Reply

Your email address will not be published. Required fields are marked *