Jeffrey Funk writes (American Affairs Journal, only one free article for non-subscribers),
In total, 2020 set a new record for the number of companies going public with little to no revenue, easily eclipsing the height of the dot-com boom of telecom companies in 2000.
He says that Internet start-ups are not undertaking any technological innovation.
Ridesharing and food delivery use the same vehicles, drivers, and roads as previous taxi and delivery services; the only major change is the replacement of dispatchers with smartphones.
Twenty years ago, I wrote a book advising Internet entrepreneurs not to seek venture capital. “Fundraising is for charities,” I sniffed. I was probably wrong–we would not have Amazon or Google today if everyone had listened to me. But if Funk is correct, then the VC industry may have gotten too big and too warped in its approach. He writes,
The poor performance of VCs and start-ups and the corresponding sense that they are mostly trend-chasing copycats are both indirect results of superficial training, and so part of the blame for these problems must fall on business schools and universities. In recent decades, business schools have dramatically increased the number of entrepreneurship programs—from about sixteen in 1970 to more than two thousand in 201426—and have often marketed these programs with vacuous hype about “entrepreneurship” and “technology.”27 A recent Stanford research paper argues that such hype about entrepreneurship has encouraged students to become entrepreneurs for the wrong reasons and without proper preparation, with universities often presenting entrepreneurship as a fun and cool lifestyle that will enable them to meet new people and do interesting things, while ignoring the reality of hard and demanding work necessary for success.
Funk sees too much focus on business strategy (“build a platform!”) and not enough on science and engineering.
technologies, not business models, enabled many of the successful start-ups of the previous generation to succeed.
Funk points out that corporate research labs pre-1970 seemed better at producing technological breakthroughs than our current system of government-funded research at universities. I do think that government money has corrupted universities.
Yes it’s true that most internet start ups are bringing no significant new technology and will end badly for the initial investors.
Even so, I am much more inclined to see this as in line with the historical norms surrounding revolutionary new technologies rather than as a historical deviation in favor of reckless entrepreneurship. My understanding is that regardless of new business school programs in entrepreneurship, overall levels of entrepreneurship are down, not up historically. I didn’t bother to find a source for that claim so maybe it’s wrong but I keep hearing there are fewer and fewer public companies to invest in.
Certainly the development in railroads in the 19th Century led to countless failed railroad companies that brought few or failed innovations to the table.
And then there is this from Wikipedia on the history of the automotive industry:
>–“Starting with Duryea in 1895, at least 1900 different companies were formed, producing over 3,000 makes of American automobiles”
So three of those had sustained success. Isn’t this what usually happens when a revolutionary technology disrupts things?
“Twenty years ago, I wrote a book advising Internet entrepreneurs not to seek venture capital. “Fundraising is for charities,” I sniffed. I was probably wrong…”
I’m curious about the hedge. What counter-evidence keeps you from saying “I was wrong”?
Regardless, it’s not clear to me how your previous advice is related to Funk’s assessment in the subsequent quoted passage. Can someone clarify this for me?
I would suggest that Funk use simple economics to explain look-alike competitors. The desire to blame business schools is wrong-headed on tw0 fronts: 1) silicon valley start-ups are dominated by engineers not business school types, and 2) business schools over the past decade have experienced plummeting enrollment.
The assertion, that many start-ups bring no material innovation other than a smart app is perplexing. Again, is Funk not willing to use any economic or business school ideas to guide his thinking? The substitution of Labor with Capital in his example is easily understood by a production function. The entrepreneurs in question posit that this change in costs coupled with a better user experience results in a better product-market fit (an easier and better customer journey). Market dynamics are further explained by competition and customer preferences.
I would ask Funk to look at the trend in Stanford majors over the past decades. He will see a material shift from economics to engineering (computer science). The students who are allegedly ill-prepared according to the study are challenging the complex and uncertain world of entrepreneurial ventures. Note the key challenge is uncertainty. When experts talk about uncertainty, only the naive and titled pay homage.
Funk may want to evaluate his thesis by looking at Nature. The search and selection mechanism is basically random. A mechanism for changing production function is bounded mutation. This change is then subject to fitness (survival). Efficiency is not Nature’s only objective measure – effectiveness also matters.
I will close by saying that innovation is a very messy process that cannot be achieved by following ivory tower proclamations. I am willing to go a step further and say following the advice of academic theorists in this area is tantamount to listing to astrologers for similar advice.
1. Where is the crisis? Not in the VC returns. At least not yet.
2. Is Uber an innovation over taxi cabs? Funk says no. But it’s an obviously superior service. It’s unclear to me why we should distain more efficient utilization of existing resources.
3. As a (perhaps mean-spirited) aside, I doubt Funk could describe the ‘back end’ technical platforms that enable most internet businesses. One might wonder if we is well qualified to assess the degree of innovation involved.
4. It does appear that the mobile phone has enabled a new *marketing channel.* The functionality of grub hub/door dash *to the consumer* is not so different from what dining in had in 2000. But market share is being directed via mobile channels. Likewise Robin Hood to Etrade. It’s the same service, but with a different customer acquisition channel.
5. In some cases mobile + logistics has enabled .com business models that failed the first time round. Chewy literally *is* pets.com.
6. That said, we are certainly seeing an inflation in assets, and companies going public earlier. Like everyone else who’s a value fundamentalist, I struggle to understand current public valuations of TSLA, bitcoin, and the like. But I don’t conclude from that sweeping statements about “start-ups.”
Chewy billionaire is now biggest 12% share owner of GameStop.
$263 – unbelievable.
https://finance.yahoo.com/quote/GME/key-statistics?p=GME
Went from less than $10 up to over 400, then down to 115-90; then more falling to 50, then 40 … then UP again. Over 100. Now over $200. My daughter thinks there are some billionaire investors having fun now – that’s as good an explanation as any others I’ve heard. (I thought it would stabilize at $30-40, very unlikely over $60, possibly back down to 20).
But I kick myself more over not investing in Elrond, the crypto Alex T suggested a month ago, now over $140 (from $63).
Imagine showing what can be done on the iphone to someone in 1989 or even 2005 and then try to convince them that there’s no fundamental innovation going on.
Uber has over 2,000 engineers, maybe it’s harder to rebuild the dispatch part of the stack than it looks? Maybe making it look SUPER EASY from a customer’s point of view is actually a massive R&D challenge?
All of these companies have been, on an off, spending billions to try to replace the driver with a computer, tackling another part of the “taxi” stack. Uber even spending money on air taxis, to try to innovate out the “roads” part. Market is kind of killing them for these and they will probably stop investing if they haven’t already but…what is this, other than what he wants?
Can’t read the article due to paywall, but this seems so lame it’s hardly worth quoting.
+1.
Also, it’s one thing to say tech companies aren’t innovating, it’s another to say new startup tech companies aren’t generating profit. Often innovation is hard to monetize and turn into a profitable company.
The poor performance of VCs and start-ups and the corresponding sense that they are mostly trend-chasing copycats are both indirect results of superficial training,
I didn’t have this feeling before, and as I look into it a bit I have it less. There’s a flood of VC money, looking for gold in start-ups. The last 20 years might have been a “VC premium” range, where there was less investment in start-ups post dot.com bubble pop, relative to other investments for same IRR (or TVPI or DPI) [older 2018 note:
https://bettereveryday.vc/how-vcs-get-measured-fbfc4d513436 ]. Maybe it was 40 years.
The Fed is printing money – much of it reaches rich investors looking to invest, and all asset classes are seeing big booms, with small corrections in individual companies.
TechCrunch 2020 Jan says VCs are tired.
https://techcrunch.com/2020/01/16/vcs-are-just-tired/
So much money that start-ups and VC targets get more cash for less equity – are pricey. Like this $5.3 billion Visa purchase of Plaid, that all VCs looked at and most said “no”.
The company had revenues in 2018 in the upper tens of millions,… a revenue multiple somewhere in the 30-50x range — extremely pricey
All are looking for premium success. Both VC investors AND entrepreneurs and even authors like Funk. Most dream, and look for their own path to riches, or at least comfort,
while ignoring the reality of hard and demanding work necessary for success.
And the BLM/ SJW movement is an attempt to get premium success with even less hard work. BLM domination of gov’t means even less efficiency of future tech, since successful science or math is “racist”. All should be outraged and furious about this.
An old Patrick Collison post about starting up in Ireland notes how important it is to see a probable, good exit before a VC will enter.
There aren’t many acquisitions happening, reducing the expected outcome of a start-up. In Silicon Valley, a failed start-up with good founders will often be bought by one of the tech giants; In Ireland, that same start-up is likely to simply fail. This matters not only because founders are less likely to want to start companies, but also because it’s harder to attract capital and to convince employees to join.
This minimization of the cost of failure is likely a huge Silicon Valley advantage in supporting entrepreneurs trying to discover what the market (customers!) is really willing to pay for.
Skim Uber’s software engineering blog:
https://eng.uber.com/h3/
https://eng.uber.com/introducing-ludwig/
https://eng.uber.com/aresdb/
That’s behind the scenes software innovation, but that’s real. They have very talented software engineers. All of those projects are given away for free as open source, so measuring innovation with profits is complicated.
On the one hand its very obvious to me that buying something through my iPhone is way more convenient then talking to a person, to the point that I would pay a premium for it.
On the other hand, its very obvious that all these delivery apps are operating at a loss (especially the food ones).
Why shouldn’t delivery apps end up like the airlines? I.E. when there is nothing to differentiate you but price why don’t you end up in a darwinian struggle with few long term margins.
The assumption seems to be that once a particular app gets dominance nobody can challenge its margins, but that didn’t work for the airlines. I can easily switch between uber eats and door dash.
its very obvious that all these delivery apps are operating at a loss
I doubt this is true, but it’s certainly not obvious to me. I’m also not looking.
Do you have any links that show this?
They are still operating at a loss but their market capitalization is anticipating large profits later on.
https://www.forbes.com/sites/aliciakelso/2020/12/10/now-that-doordash-is-public-heres-how-it-plans-to-be-profitable/?sh=58fc38bd6472
DoorDash operating profit the past 3 calendar years: -210M, -616M, -436M USD. Uber is also very negative but since it’s not only food delivery I will not quote the numbers.
I will routinely see situations where Ubereats is giving me 40%+ off my meal when their entire revenue before expenses is 30% if I’m not mistaken. That means that are losing on all those transactions in order to inflate transactions. If Ubereats stops basically giving me free meals, I will either go to a competitor that gives me free meals or reduce the amount of food I order.
It’s difficult to tell if I’m a “real” customer of their service properly priced to make a profit.
Amoretti XV: Ye tradefull Merchants that with weary toyle
– Edmund Spenser
Ye tradefull Merchants that with weary toyle,
Do seeke most pretious things to make your gain:
And both the Indias of their treasures spoile,
What needeth you to seeke so farre in vaine?
For loe my love doth in her selfe containe
All this worlds riches that may farre be found,
If Saphyres, loe hir eies be Saphyres plaine,
If Rubies, loe hir lips be Rubies sound:
If Pearles, hir teeth be pearles both pure and round;
If Yvorie, her forhead yvory weene;
If Gold, her locks are finest gold on ground;
As a former tech materials startup engineer, I’ll say that explosive growth in one area can crowd out innovation in other areas. The ceramic nanocomposites and liquid semiconductors I’ve worked on were more significant innovations, technically, than yet another online storefront or cell phone game, but the path to modest profit was by nature long and uncharted. The capital markets of the time found Petstore.com far more appealing.
It’s a good point. Impressionistic analysis of innovation is prone to both type-1 and type-2 errors.
People overrate a lot of famous and salient stuff which are hardly big deals in terms of actual technological achievement and quite often the enhanced utility is misattributed to the tech itself instead of, for example, a change in the de facto regulatory environment. “Low-hanging forbidden fruit”.
When someone tries to see if it’s really still forbidden, and, whatever the law books might say, as effective reality it’s not, then everybody gets to munch on the apples. Uber and AirBnB were illegal from the start in most places, and technically still are, whether or not those laws are actually being enforced.
On the other hand, people under-rate innovation which is boring and technical and invisible to non-specialists, but which have a huge spillover effect. There are a lot of technologies which all face a similar kind of fundamental limitation or bottleneck in the cost or capability of some function or piece of equipment. Countless other things tend to stay expensive and stagnant until someone can figure out what to do about it.
But, if someone has a genius breakthrough for that one issue or function, then the ripple effects are enormous, and a flourishing of innovation is enabled by a lot of quick, easy, and objective unimpressive ‘innovation’ is the consequence, which is lot of people in lots of fields simply swapping out the old part for the new breakthrough.
In the past, this happened with motors, both internal combustion and electrical. Inventing and developing a process to make good motors is hard and took a long time even after the theory and principles involved were well understood. But once you have them, hey put that motor on wheels and a plow and now you’ve ‘invented’ the farm tractor. Put in on tracks and put a big scraper in front and now you’ve ‘invented’ the bulldozer. Put some propellers on it … and so forth. Those periods are so disruptive because one generally-applicable breakthrough will quickly revolutionize lots of activities.
Something like this ‘recently’ happened in magnets, and of course, got very little press. But engineers know how many things depend on magnets and how many constraints are imposed by it being hard to generate strong fields in compact spaces without providing impracticable amounts of power and cooling (e.g., trying to do it running electricity in coils of ultra-cold superconducting wire).
If we actually get some progress in nuclear fusion ‘soon’, it will probably be the result of some of these advances in certain kinds of magnets.
But there are many more mundane applications too, using what are now less exotic magnet materials, like Neodymium. MRI is one example, but there are lots of others. Over two decades ago, a certain kind of research with which I am familiar had to use a piece of equipment the size of a car, which had to kept in a room as big a Jiffy Lube garage, and which needed tons of electricity and liquid helium to run. It cost well over a million dollars, probably nearly as much per year in maintenance and operational costs, and could only be operated by a tiny number of the most trusted and experienced professionals. Almost all of that was necessary just to make the desired magnetic field.
Fast forward today, and there are now comparably tiny, permanent magnets which produce even stronger, better fields, without cooling, without huge amount of electricity. All of a sudden, a piece of equipment that does an even better job is 100 times cheaper, required almost no maintenance, can be used by anybody safely and easily, and can plug into a regular outlet, or even run on power-tool batteries! They are small enough to fit on tabletops, and even in a backpack for use in the field!
These machines and related pieces of analytical equipment used to be enormous bottlenecks to all kinds of scientific research and technological development. The fact that it is ‘suddenly’ possible to have tons of them has dramatically improved the throughput of a lot of activity that is the background foundation for many other advances which manifest in biochemistry with medical applications, for example, in many of the hopeful bright spots in this largely depressingly revealing pandemic.
But these are not ‘biochemical’ advances per se, in terms of new kinds of chemicals or synthesis pathways, etc. They are mostly putting a scraper on a motor and saying you invented the bulldozer. That is, they are what is enabled by having lots of analytical devices which are suddenly *100 times* cheaper, faster, and easier to operate, which in turn is also just a ‘bulldozer’ of sorts, because those were also enabled by one really important breakthrough in magnets. Which, it must be said, might just be a one-off we won’t get again.