« The (near) inevitability, and who and when, of Helicopter Money | Main | Time to raise the gas tax? »


Feed You can follow this conversation by subscribing to the comment feed for this post.

Yep. Statscan can measure producers surplus somewhat acurately (profits) is incapable of calculating consumer surplus as it is currently unknowable. (As is the value of extra weeks of vacations bu Europeans whom Americans deem poorer in view of GDP data.)

Jacques Rene: I'm not sure it can measure producer surplus either. Suppose our jobs got nicer (or nastier) over time, but we produced the same amount of goods and services. The GDP data wouldn't change. We don't send small boys up chimneys any more.

(And I'm not sure if my broken spring was caused by Quebec roads, road debris, my driving, or simply old age metal fatigue, which opens up a whole new can of worms in measuring GDP, because we should maybe count spare parts as an intermediate good.)

Hmmmm. Sounds vaguely similar to Brad Delong's recent musings on Star Trekonomics.

I must say, I think the IT part of the process is probably the least interesting part. I think of the internet as a giant real time inventory that everybody can browse through. Like a big phone book or catalogue, organisable in different categories. Then there's the payment stuff that's attached to it. That's all intellectually interesting, but not logsitically. The logistics on the ground I are much more amazing, in an archaic way (here's a young(er) person speaking). Someone had to physically be there in no time and pick up the bloody coil and then drive / fly it to you. All for, what, 4$? Sending bits and bytes around the world at the speed of light is easy, but having the delivery end keep up with the digital world takes an army of cheap, 24/7 drivers, a network of logistic centres around the country and cheap oil. One tight ship!

I also imagine the difference between then and now does show up in the GDP / capita data. The fact that you received your coil within 2 days means you were up and ready to work much quicker than you would have been 20 years ago. The extra work you were able to do in those 2 weeks added to GDP. Sourcing the coil via phone and what not 20 years ago would also have taken up many more work hours by people other than yourself, meaning productivity was much lower. And if you look at output per hour worked and absolute hours worked that should give you a sense of whether time not spent working is idleness or vacation time. For those that actually have work, that is. The question is, is there a trade off between more vacation time for the employed and overall unemployment? Can a lump of labour be divided equally or does work create more work? And who decides what's best for whom? And is higher productivity always good? Is working slowly a kind of permanent vacation?

Chris: Does Brad know as little about Star Trek as I know about logistics?

Oliver: "The logistics on the ground I are much more amazing, in an archaic way (here's a young(er) person speaking)."

Hmmm. Me too. I still don't see how they got $50 worth of dust covers from Chicago to Gatineau in a little over 24 hours. They estimated one week. (Unless they found some closer). And the physical transportation can't have been that different from (say) 50 years ago -- presumably trucks and planes.

On increasing measured productivity: yes, if I had been (say) a taxi driver, and my taxi was back on the road quicker.

Nick - always like your car posts!

Random thought: Hayek's basic thesis was that markets were awesome because they were really great at transmitting information.

But in your story, the information isn't transmitted through the market - the Mazda dealer is getting the info from the internal Mazda computer system. Kind of like a centrally planned economy, only a bit more efficient.

What happens to economies and to economic organization when information can be transmitted super-efficiently without price signals?

Frances: thanks! The way I look at it, you need the Hayek story, because without prices nobody would know relative scarcity or have the incentive to act on that relative scarcity. But you also need the potential buyer and seller to be able to find each other.

The Mazda parts guys was (presumably) just using Mazda's internal system. But the non-Mazda parts store was presumably connected to multiple networks from different suppliers. And then there's Ebay.

Also - the existence of the networked inventory system means that the inventory itself, although fragmented like grains of sand on a beach, encourages the maintenance and increase of that inventory.

Your spring supplier might reasonably have decided years ago to sell his old springs for scrap metal, or for an art installation. But because of the universal inventory, there was a reasonable hope of that spring being needed.

Better yet, this encourages the practice of sentimental inventory. Someone who respects books, or machinery, or craftsmanship, will have been pressured in the past to finally, reluctantly dispose of items cluttering their basement. But now, even though a sale of one of these items might never remotely recoup the cost of storage, the exchange when that item goes to a new owner can be a source of deep pleasure and satisfaction to the seller and the new buyer both -- a collaborative pleasure that will outlast any purely monetary considerations.

@ Frances
The way I understood the Hayek paper is that you want people who make ad hoc decisions in their own responsibility all along the supply chain because only they can judge whether the prices being charged correspond to actual, on the ground scarcities. And if not, only they are in a position to find more efficient alternatives. The whole paper is a song of praise for the entrepreneurial tradesman.

The internet, because it has eyes everywhere, somewhat eliminates the need for these intermediaries. It may be organised hierarchically, but it connects random people, i.e. creates an efficient market.

I would add though that the logistics companies that make the whole internet shopping experience work, say UPS, are probably run more like the military than a network of free entrepreneurs. Unless Amazon starts working with Uber to get their stuff delivered, that is.

The next thing you know, Nick (who is now also an Uber driver because his wage at Uni was cut) will get paid 2$ for picking up his own spare part at the Amazon logistics centre and delivering it to himself.

Noni: "Your spring supplier might reasonably have decided years ago to sell his old springs for scrap metal, or for an art installation. But because of the universal inventory, there was a reasonable hope of that spring being needed."

Yes. Though I would rephrase it to read: "....there was a reasonable hope that he would learn who it was that needed that spring."

And on the other side of the market, I keep my old MX6, because I have a reasonable hope of finding someone who has springs and other spare parts. Sort of like a multiplier.

Nick - let me Coase your Hayek. You wrote:

"you need the Hayek story, because without prices nobody would know relative scarcity or have the incentive to act on that relative scarcity"

Sure, but think about Coase's theory of the firm - transaction costs - what the internet is doing is radically changing transaction costs both within and outside the firm, and the end result may be an economy which is either radically more market-y or radically less market-y than was once the case.

Most people would say...uber...airbnb...sharing economy...contracting out....internet is expanding the scope of the market. But your story - Mazda internal network finds parts in a super-efficient way - contracts the scope of the market.

Your story is, if I understand correctly, about nominal v. measured v. real GDP. I want to tweak the story partly because I don't find GDP measurement all that interesting, and partly because the scope of the firm matters for the measurement of GDP. If you buy your springs from the guy at the U-pick, the money stays in Canada and contributes to Canadian GDP. If your springs are brought into Canada by one bit of Mazda talking to another bit of Mazda, the allocation of spring costs between the CAnadian Mazda and the American Mazda is essentially an accounting fiction. And that matters for Canada's GDP - what does Mazda Canada pay Mazda US for those springs? Pick a number - it's not a market transaction, it is to a greater or lesser extent a work of fiction.

Frances: Hmmm. I think you are right to bring Coase in. The Mazda network is internal to the firm; but the one between my local parts store and the various suppliers is a network between firms.

"Your story is, if I understand correctly, about nominal v. measured v. real GDP."

Well, that bit about GDP was more of an afterthought. Just another example where (real) GDP doesn't correctly measure changes in welfare.

(Just finished replacing the rear springs, struts, and strut mounts. Monroe only had one rear strut left, but they found two KYBs. Ordered them Saturday, and they came in Tuesday morning. It has raised the ride height an inch or two; not exactly "donked", but definitely not "stanced".)

Nick, Brad was musing about how we are 95% there to a replicator. You buying your car parts so easily is 95% of the way there too.

I am not sure I understand in your first comment when you say” What happens to economies and to economic organization when information can be transmitted super-efficiently without price signals?”. It is the prices themselves that are the information. Having a more efficient communication mechanism of those prices (and of the quantities offered), certainly make transactions more efficient. So information ABOUT prices is great, but it is the prices themselves (and relative prices at that) that are the economic information or “knowledge” provided by markets. Even if the Mazda network is internal, there are still external market prices available for comparison.
The extreme form of efficient price mechanisms are things like the Toronto Stock exchange, or EBAY where information about prices and quantities is made available nearly instantaneously, and the cool thing about Hayek is that there is even a market and hence price information for those very market mechanisms themselves! You can list your stock on a variety of exchanges, NYSE, NASDAQ.... or use Craigslist vs. EBAY for Nick’s car parts based on among other things, the price of using those systems.

I can relate to this, and have spent some time thinking about it.

I am interested in bicycles, and the existence of the internet makes the finding and procurement of old parts possible.

Two observations: one, that there is a large increase in welfare without much impact on GDP not just because there is a matching of buyer and seller that might not have occurred at all, even if the price doesn't fully reflect the consumer surplus that might arise, but also because a large portion of the sales probably happen without being recorded as sales, because they are private transactions that are not recorded as income. Also, further complicating that issue is that many of these sales are international.

Two, the increase in efficiency has reduced the serendipity, and the ability of people to find a stash of something somewhere for much less than the market price. For bicycles, there are collectors all over the world who troll message boards, eBay, Craigslist etc... searching for that perfect part. The trolling doesn't even have to be manual - it can be automated so that one is alerted when certain things come up for sale, no matter where in the world it is. This allows full value to be had for some of the items, hopefully motivating more hoarders to sell of their stash, and for more people to preserve stuff (like car springs or bicycle parts that otherwise would be pitched, or if there is luck, found by serendipity and sold for much less than market value).

One other thing - I agree that the logistics are a big part of it, but I don't think that is as transformative as the ability to find things through reducing search costs. I used to (in the pre and early internet days, when I was into working on cars) order stuff through mail order. It was not bad, as long as you had the right catalogue (and early on with the internet before peer to peer sales really picked up, or there was a lot of ecommerce). Sure, things get to you a lot quicker now (even in Canada, which is still painful compared to being in the US for delivery and online shopping), but for many of these items, it is the savings on the search cost that is the biggest issue, especially where in many cases you couldn't even get the part because buyer and seller couldn't be matched. Sure, it is impressive that Nick got the springs in a couple of days, but that is probably less important than the fact he could get them at all. Having your car up on blocks for a couple of extra days is way less costly than having to scrap it because you can't find the right springs.

I remember some economist saying that after decades of investment in the IT industry it has pretty much come up neutral in terms of any gains to business from adopting modern databases, communications, etc. I didn't believe him for a moment, otherwise there would be a lot more firms that don't bother to use this technology, but when you look almost all of them do. As you say, the young buyers now just expect it. Whether this is inside the company, using internal processes, or whether this is out on the open Internet using eBay or whatever, makes no difference. If you don't use the technology, pretty soon you will be out of the game.

As for the gains to society as a whole, going down that global utility function rabbit hole... I consider interpersonal utility comparison and especially inter-temporal interpersonal utility comparison to be somewhat unscientific at the best of times. GDP is a very rough and ready measure. It should just be accepted that there's a bunch of things poorly measured.

I notice that others have mentioned Coase and transaction costs, I'd like to point out that the actual transaction part of transaction costs have been low for a very long time (i.e. the overhead required to make a payment or transfer money). This is only the tip of the iceberg, in as much as Coase (rightly or wrongly) dumped just about *EVERYTHING* into that "transaction cost" bucket. For example, negotiating an agreement counts as a transaction cost, but Internet is not better than telephone in this respect. Navigating past legal barriers and licensing count as transaction costs, but lawyers are still lawyers, and if anything more of these barriers exist now than ever before. Then there's the trust question: if you buy on eBay you might get a defective part, even if eventually you do get your money back it's still tedious and time wasting. Having an efficient way to give money to a scam artist, does not solve the deeper problem that scam artists do seem to thrive.

Firms actually solve a great many issues that are not really related to transaction costs as such; but for the purpose of a good theory you can lump them in with transaction costs (and fool some people by doing that). The frictional component related to search costs has definitely been reduced by IT, I would agree with that much.

The comments to this entry are closed.

Search this site

  • Google

Blog powered by Typepad