Looks like Uber is better on every metric, total volume, growth, revenue, the only area they are behind is margin where they are taking less from their drivers than Lyft is.
While Uber looks healthier especially given it's size, Lyft's reception after the IPO was quite negative, with the stock falling 30% immediately in the days that followed.
Even with better economics and larger scale the larger question looms, which is how do these businesses turn profitable.
Are they subsidizing a market and creating it by charging less, or will they reach a certain scale and be able to cut back on certain expenses which will give them profitability.
I don't think Amazon is a fair comparison here, because they were investing in infrastructure, building out global logistics, which is different from Uber and Lyft. They should be much more profitable because they don't have to invest in that. They don't know warehouses and logistics, and everything else that e-commerce required.
Also in the case of Amazon the supposed profits would be reinvested in the business with large CapEx spend, here it seems that the negative margins are being spent on sales and marketing, which isn't the same as what Amazon was doing when it was treading in the negative for a decade.
Looks like Uber is better on every metric, total volume, growth, revenue, the only area they are behind is margin where they are taking less from their drivers than Lyft is.
We lose money on every sale but make it up in volume is supposed to be a joke, not a business strategy!
I've reported several fraudulent trips; trips where the driver accepts my fare and pretend he's actually picked me up and then ends the trip after some time. When I requested a refund, Uber happily credits my account, not my credit card that they billed.
Seems like some fancy accounting tricks that bump up numbers are being ignored too.
Crediting my account isn't a refund. I never requested a credit. So I stand by my assertion that it is fraud.
The same goes for having to meticulously scour my trip logs for where the driver didn't end the ride when I exited the vehicle, or other such shenanigans.
I don't think GP's comment was challenging your view but was more like advice ("if you complain enough to support, you'll get a refund instead of a credit").
You should just ask for refund. I've the similar story, Uber suggested credit. But immediately issued refund once I declined credit and asked to refund money back to my credit card.
I don't know about Lyft but Uber says they lost money in Q4 2018, hopefully not in each sale but definitely in aggregate.
The money the got from customers for rides / food delivery didn’t cover the money they paid to the drivers / restaurants providing the service.
Edit: Ok, looking at the numbers more carefully they are still be positive after paying drivers / restaurants. But they lose money when the operating, marketing and administrative costs directly attributable to the sale are included.
I think the idea is those costs don’t scale linearly with the number of rides, so if you can increase the volume of rides which each have a small profit and hold the other operational costs constant or at least less than linear, you eventually get to profitability. That’s the theory anyway.
Sure, but right now they are losing money with every sale. Using their own "Core Platform Contribution Margin [which] demonstrates the margin that we generate after direct expenses. We believe that Core Platform Contribution Margin is a useful indicator of the economics of our Core Platform, as it does not include indirect unallocated research and development and general and administrative expenses (including expenses for ATG and Other Technology Programs)."
According to Aswath Damodaran [1], Professor at NYU and an expert at valuing companies, if he had to invest between Uber and Lyft, he will choose the latter.
Who are we kidding? Uber, already a gigantic loss machine, continues to invest in loss-making ventures like Food Delivery and E-Scooters. Uber has spread itself to so many areas and markets that I can't see how it'll ever be profitable without making huge cutbacks. The volume and growth are of little significance when they are failing miserably to contain their losses.
Compounding it is the fact that there's nothing sticky about any of their business ventures. Customers will choose the cheapest option, so they can never go beyond a certain range of prices. Even if Uber does become profitable, I highly doubt it ever will join the leagues of Apple, Google or Facebook.
I don't know anything about the man or his success record, but as an academic he may not wish to be accused of conflict of interest by promoting companies where he has a personal financial stake.
Academics (pure academics) don't have a good track record predicting valuations/prices, period.
The people who actually have the ability or to predict valuations/prices (better than the market) are busy making millions, if not billions, of dollars, and they most certainly aren't going to tell you about their predictions/models. If they do, it's after they've already invested or shorted the asset.
This isn't quite true and you've pretty much identified the reason why. Academics are not investors and their objective isn't to profit but to publish their research.
Academics identify a pattern that generates superior risk-adjusted returns and then publish their findings - this then leads to the anomaly disappearing as investors trade away the alpha. [0] The track record of an academic can therefore only be meaningfully discussed in terms of how well their model performs in back-testing. Saying they don't have a good track record misses this point.
Perhaps there are successful academic investors who achieve alpha and don't publish their research, however they wouldn't show up in any meta analysis.
I’m not aware of Perelman publishing a “a pattern that generates superior risk-adjusted returns".
That type of information is worth millions, and there are many investment firms offering very lucrative positions to those that can identify a pattern that generates superior risk-adjusted returns.
Outside of very rare outliers, it doesn’t seem reasonable to assume that information that will generate outsized returns will be publicly available.
Because if your chosen career is an academic that's what you do. You research, write and publish. You could probably make more money by selling your ideas in private industry but that's an entire career pivot, it's like a junkyard becoming a cheap used car lot. It's a lot of work and comes with risk. If you're tenured or close to it you should probably just publish. Sure you could try and sell your ideas on the side but that's not going to endear you to other people in academia.
It's possible to express an opinion on the relative value of two stocks while remaining neutral on their correlated returns, by going long one/short the other in the appropriate ratio.
A fair valuation may not be the same as market valuation since you can always bid for a stock high if there's a greater fool to buy it off you. This is what bubbles are made of - and once the music stops valuations will return to the fair valuation derived from multiples of revenue or asset values.
Cool window into an interesting epistemological model. Considering the null hypothesis: "Damodaran is no better at predicting the market than any other person.", what would be evidence that rejects it?
Presumably not being in concordance with the market is insufficient evidence based off this and sibling comments. I wonder if there is a duration where this will even work.
Does there exist any evidence that would lead one to fail to reject the null hypothesis? Or would this model always reject?
Certainly, our current conclusion must be that Damodaran has poor predictive power over the short-term.
Before I have seen the musings of a fund manager posted, predicting the next downturn. Apparently he’s been doing this since before the last downturn. His fund is barely above flat since inception in the early 2000s, even though the S&P has more than doubled. After pointing out all that, I still got some respondents who claimed that active funds are known to underperform during good times and outperform in bear markets. Perhaps it is true but I remember feeling skeptical.
I guess there is no evidence that will convince everyone that someone is a bad money manager or prognosticator. That’s why, despite my general discomfort with Nassim Taleb, some of what he said in Skin in the Game really seems true. People can believe what they want, but beliefs don’t matter much if they aren’t confirmed, and the only meaningful way to confirm or reject market beliefs is via investment results.
Suppose you have someone that says "amazon is overvalued, there's no way they are worth more than $400B" over and over again for years as Amazon goes from $600B to $900B. I'd still keep an open mind as long as the market in general was irrationally exuberant over that period. It's only once the market as whole returned to sanity and most stocks went down, but amazon continued to raise that I'd say that guy probably doesn't know what he's talking about.
"there's nothing sticky about any of their business ventures"
Their critical mass of 'drivers' and their brand are hugely sticky, and massive barriers to entry.
Once Lyft and Uber have made sure that they are the only players, prices will inch up as they de-facto collude together, just like supermarket chains or anything else.
If a new entrant in a city tries to challenge them, they'll price way below market value until that challenger is gone, and then bring prices back up.
Wins and losses will be on a city-by-city basis, so it's possible that some city, somewhere has a 3rd party that wins out. Like maybe in Texas, due to whatever framework some other entity 'won' and came 1st (i.e. 'in Texas we proudly use 'Whatadriver'!), then it's possible for 3rd parties to exist, but the same rules will apply.
This market will settle down, and both Lyft and Uber will cut all of their stupid projects, and then they'll settle into oligarchy type pricing and profits.
As far as valuation, that's difficult to tell, but I think they have very viable business models that are not going away.
Facebook and Twitter could dissapear as quickly as they came, but Uber/Lyft found a back door into a very ancient market and they have their claws in their now, so they are not going away. They'll be around. It's just a matter of valuation at this point.
Most of the ubers and lyfts I've been in had both apps running. If a third app comes in with decent driver incentives, and can convince google maps to show them, they'll be able to get passengers and drivers. And, as you said, they only need to do it city by city.
I will say, when i needed a ride in a more rural area near Seattle, I did go right to Uber, and that driver wasn't on Lyft (but was considering it), so there is some stickiness, but I don't think it's much.
I see your point, but there has been a zillion attempts to do the Lyft/Uber thing and tons of money has already tried. The pie is just to big to ignore. So if hasn't happened already, it's not likely going to happen.
So yes, maybe a 'serious competitor' could try in some city, and as you say maybe 'word would travel fast' and maybe it might be possible to get a critical mass of drivers ... but then Uber/Lyft will just undercut them in a war of attrition.
To your mini-case: maybe there was a few other 3rd parties in existence, but you possibly never even heard of them. And even if the drivers sign up ... without the demand well there's no game anyhow.
It's been commoditized, so the advantage now is scale, brand etc. which the incumbents have, and they have money.
That market is done for now, until something comes from a different angle: new tech, new regulation, public transport shift, self driving cars etc.
You can't fight a subsidy war with uber/lyft now while they're still in growth mode. Once investors start demanding profits, and the prices go up, that's the time to come in and spend your money driving people around. Especially if you hear about others doing the same in other cities, or when some of the big loans are coming due.
It's a commodity, brand doesn't matter. Scale across cities doesn't really help that much. Money helps, but public market investors don't like to see it being spent quarter after quarter.
If food deliveries are made using the same driver fleet as the transport business then it is actually essential to squeeze more revenue per driver hours. Having the fleet there only to serve transport is inefficient because driver time is fungible between the businesses.
Why do people compare Uber to Apple, Google or Facebook? Why do people talk about them as if they were a tech company?
It's a taxi company. They have a web site and a mobile application, but so does several retailers as well as pretty much every bank out there. And no one is accusing them of being tech companies, or expect their growth to mirror that of Google or Facebook.
If you just look at the business of Uber, it's clear their current business does not inspire confidence that they will ever be profitable, much less deserving of Google-level valuations.
Sure, they may be able to become successful, but that would be on the back of a different business model. But then you'd have to justify why Uber would be able to execute on this as-of-yet-unknown business model better than any competitor.
Uber Eats revenue was $1.5 billion in 2018, representing an increase of 149% from the $0.6 billion in revenue the food delivery service generated in 2017.
Honestly I don't understand how anyone could be thinking that investing in Lyft's stocks is a good idea when they have a competitor like Uber. It's like investing in eBay when there is Amazon or investing in homestay when there is Airbnb.
Further to that point and to their credit, ebay smashed Amazon in auctions.[1]
Auctions was one of a dozen or so large failures on the part of Amazon to find a profit center. eBay wiped the floor with them and Amazon gave up relatively quickly. Since then, eBay has generated over $15 billion in profit in auctions, a figure Amazon struggles to match in its entire retail history despite the scale difference.
All these years later, eBay is generating ~$2.3 billion in operating income and their auction business is under no serious competitive threat (it is a slow growth business however, basically fully saturated). Amazon's huge retail business is only barely able to match the profit of eBay's auction business.
The ultimate value in Amazon of course wasn't retail. For eBay, the value is actually in the ecommerce.
eBay has indeed found an amazing niche. I've been gladly forking what one would consider an outrageous fee for auctioning off one's items, simply because the stuff sells so predictably. I regularly list stuff that one would consider junk and make a buck. Just auctioned off a decade old Asus eee PC for $50 (it still works though). Im so okay giving ebay 5 bucks for this ability! I made a few dollars, a working laptop found some home, and i threw less into the trash!
> I made a few dollars, a working laptop found some home, and i threw less into the trash!
I love this about auctions, it seems to preserve value. Rather than e-waste going for scrap it might go to a collector, upcycler, charity, or specialist recycler.
Since then, eBay has generated over $15 billion in profit in auctions
Source? eBay has squashed most of its own auction activity in favor of fixed-price items. Participation in auctions has been spiraling downward for years, resulting in underpriced items, resulting in sellers abandoning auctions altogether.
I suspect part of the decline of auctions is that they're risky for low-velocity products.
Example: your item is worth $200 on the fair market, but might only sell one or two a month. There are simply not that many people chasing, say, a Commodore 4040 disc drive or a high-grade 1898 half-crown. You can park it on a Buy it Now listing forever until you get $200. If you try to auction starting at $200, or with a $200 reserve, people will pass because if they're not getting a deal, they'd better at least get immediate gratification rather than waiting 10 days for the auction to close. You can list at $100 or $1 at auction and get interest, but then you're praying people get interested enough to push it back to $200. Buy it now is the safest route for that.
I tend to use eBay these days as "one step above Banggood/AliExpress/etc." They have many of the same items but there will usually be a few sellers offering it dispatched from the US, letting me get it in 3 days instead of 25, for a couple dollars more.
> Lyft's reception after the IPO was quite negative, with the stock falling 30% immediately in the days that followed
Pretty much everyone who invested before the IPO is up on their money. Lyft’s is not a story of a crap company going public. It’s a story of a crap IPO at the wrong price. JPMorgan priced the stock too high, didn’t adequately surveille for bad short selling, and generally numptied up a straightforward process.
Amazon is a decent comparison. Not because of a similar capital investment requirement though (which as you noted in Amazon's case went into things like warehouses). Rather, because the original business isn't where the great value turned out to be.
Amazon has a business that is worth an optimistic $150 billion if it were only traditional retail, without AWS and their relatively new ad business. That's assuming a richer valuation than their peers due to expectations of growth.
The real value to Uber is not going to be in the ride hailing. It's going to be somewhere else, spawned out of the original business, riding on the back of the huge network they've put together in some manner. They'll find something else to monetize with richer margins, or they'll never justify a $100+ billion valuation. With everything they might be able to touch with their scale, I'd bet on them finding enough other business opportunities over time (that said, I also wouldn't be a buyer anywhere near the IPO valuation). The $10b IPO will buy them another likely five or six years of additional cash burn. And if they're smart, in the first year after the IPO, they'll dilute again and go get another $5 billion or so. With their need to find another business, their AWS, they can't have enough capital.
The stock being down doesn't mean that the IPO was quite negative. It just means that the price for the stock went down. If Twitter stocks were down after IPO, would that mean that buying FB stocks is a bad investment?
The stock being down in the months after an IPO generally means the company raised on good terms as they were able to get more than market value for the equity they sold. That’s unrelated to whether it’s fundamentally a good company.
Basically I always suspect every IPO to have a period of rising stock prices, and eventually it will dwindle. That's just the way the market works. You can't expect prices to rise indefinitely. Some corrections come sooner than others. I find it much more healthy for Lyft to not have an extreme correction at least.
Note I'm not invested in Lyft or any related company (Uber), I'm just a humble novice investor.
There's still a difference between "stocks are down because the IPO was way overvalued but the company makes a healthy profit" (FB) and "stocks are down because the IPO was a little optimistic for a company that's bleeding money and has a bigger competitor".
> Even with better economics and larger scale the larger question looms, which is how do these businesses turn profitable.
With self driving cars much further away than previously thought, this is the real danger for a company like Uber, who is bleeding money through their standard taxi-like services, quarter after quarter.
Public cloud is about $300b per year right now. Assume amazon gets 66% long term, so $200b (this is unlikely for many reasons...) Cisco gets about $50b of revenue currently, and is worth ~250b. Extrapolating gets $1T. (I certainly don’t believe the above analysis; it’s terrible in multiple ways. For one thing, it ignores profit margins.)
My buddy works for a law firm and was telling me that they often represent technology companies when they IPO. It turns out there’s an intense amount of research that goes into the S-1 process. The law firm makes a huge spreadsheet, with each row representing a sentence from the S-1, along with evidence supporting the truth of that sentence. They do this because, if the price goes down after the IPO, there’s inevitably a lawsuit, and the company needs to justify everything it said in the IPO.
If lawsuits are such a huge issue, why don't companies limit their IPO to, say, 1% at very low valuations and then follow with a larger ATM offering when stock prices reach a better level to get more cash? Because the banks don't let them?
Who would get to buy that 1%? If there’s any chance of that even remotely looking like “friends of the company owners” or “friends of the banks selling the stock”, people will sue.
Even if you make it a lottery, you will have to make sure everybody who would want to buy at the low price gets equal opportunity to do so.
Finally, I would think the tax department could see selling at a price that you know is significantly below market value as a gift that is taxable.
I'm not sure why more companies don't choose to go the direct listing route (like Spotify). Set a reference price, but allow the employees and market decide.
Tech companies especially have enough employees that own shares and want to sell that liquidity should not be a big issue. This allows the price to be tested while slowly adding volume (as price stabilizes, more employees willing to sell).
Agree and the analysis being done by what I believe is a 'numbers guy'(from his CV) indicates that he will try to draw conclusions from numbers and that's a big if that assumes the numbers are equal or even accurate in how or who at each company is doing the measuring.[1] In fact that the numbers are different in the ways they are seems to be on it's face an indication that they can't be used to compare.
[1] A very small example might be 0 to 60 acceleration in cars. If done by the manufacturing what is the weight of the 'jockey'(driver)? We never know that or I have never seen it (maybe a magazine when testing uses the same driver but I am not talking about that). Yet the weight of the driver does matter. I know that when I drive my high powered but small 'sports car' the car definitely accelerates slower when my wife is in the passenger seat. So an extra X pounds would make a difference in the comparison stats.
It's weird because he calls it out in the comparison as a potential explanation for the gap in average booking per ride:
"This gap is likely also due to Uber’s heavier international presence (where they now generate 52% of their bookings). It would be interesting to see this data on a country-by-country basis (or, more importantly, a market-by-market one as well)."
yet still proceeds to draw this questionable conclusion:
"Lyft has made impressive progress at increasing the value of rides on its platform and increasing the share of transactions it gets."
I'm not so sure about that. Yes, the baseline numbers will be very different, but the economics of ride share are pretty much the same anywhere: there's an existing market price for taxis, which is more or less inflated everywhere due to regulation and inefficiency, and Uber/Lyft/whoever can provide a similar or better service at a profit... except that there's competition fueled by a bonfire of VC money pushing the margin deep into the red.
Think about factors such as: cost of vehicle ownership, cost of fuel, minimum wage, social policies, unemployment, public transport infrastructure, population density, specific ride sharing regulation, traffic congestion.
All of these directly effect the unit economics of ride sharing.
You're missing my point. All those factors are real, but they set the unit economics of taxi service, which in turn sets the baseline for Uber/Lyft pricing.
So if the market price for taxi service in a city is X, already accounting for all those factors, Uber/Lyft should be able to deliver the same or better service for (say) 0.9X, again accounting for the factors but adding in efficiency gains from bypassing parasitical taxi medallion owners, not needing dedicated vehicles, having much easier ordering, etc. There will be some variance in how big that factor is (eg. NYC's $1 million medallions gave Uber a really juicy margin to exploit), but overall, if a city can sustain taxi service, it should be able to sustain Uber/Lyft too.
overall, if a city can sustain taxi service, it should be able to sustain Uber/Lyft too
So if a city wants to kill Uber/Lyft, all they need to do is to attack the rideshare margins directly. (Not that they really should. Uber/Lyft is probably a civic good.)
These 'per driver' stats are not worth comparing when one company operates globally and the other operates in US+Canada. Nothing can really be concluded from them.
The non US markets would be expected to mostly drag down Uber's numbers though, for the most part they are significantly better. Its surprising that Uber didnt at least break out US vs non-US markets, maybe the story isnt really that good.
the buried lede is lyft's opex spike right before their ipo. typically you'd try to subdue expenses to make the profitability numbers look better, unless you have other problems to fix that you can spend your way out of.
(taking these graphs as truth) lyft's gross margins were flagging, so it seems they goosed gross margins with opex, like marketing (likely promos) and better customer service (e.g., refunds). they spent their way to higher gross margins so they could look better than uber for ipo.
but as the rest of the analysis points out, lyft is seriously behind uber in the race to overall profitability despite higher per ride take rates. they're still a risky investment in comparison.
I can see some of these stats in action. Where I live (Bay Area) Lyft is always more expensive (I typically check them both), sometimes by a lot. While in Manhattan I find Lyft cheaper about a third of the time. So no surprise I take more uber rides.
I have more incentive to stick with Uber now I’ve read this as the drivers make more (and typically the same driver does both anyway)
Drivers use Lyft for a reason; they may have a greater bonus payment that balances out the higher commission to Lyft. As a rider, you should use whichever platform will get you to your destination the soonest and for the least amount of money.
> Drivers use Lyft for a reason; they may have a greater bonus payment that balances out the higher commission to Lyft
Drivers use Lyft because riders use Lyft, and it's better to drive for either at a given time than neither. It doesn't mean Lyft pays equally or higher.
There’s a lot of price discrimination going on, as I always see different prices in the apps from my friends, and due to me using lyft more historically (probably?) I get lower lyft-line rates about 70% of the time.
I have yet to find a time or location where Uber wasn't more than Lyft when I needed a ride. Sometimes it's really egregious (50% more)--maybe during surge times?
It is entirely possible for two companies to be the same but also different in how they see their market and revenues. So, without having a deeper understanding of reading financials, comparisons drawn solely based on numbers are flawed.
For example,
Uber says:
> We derive our revenue principally from service fees paid by our Driver and restaurant partners for the use of our platform in connection with our Ridesharing products and Uber Eats offering provided by our partners to end-users. Our sole performance obligation in the transaction is to connect partners with end-users to facilitate the completion of a successful Ridesharing trip or Uber Eats meal delivery. Because end-users access our platform for free and we have no performance obligation to end-users, end-users are not our customers.
By making this distinction, Uber might be on a path of trying to maximize it's driver/restaurant unit economics and not on the per user metric.
While Lyft also thinks of drivers as their end customer but doesn't clarify the position of end-users like Uber:
> We provide a service to drivers to complete a successful transportation service for riders. This service includes on-demand lead generation that assists drivers to find, receive and fulfill on-demand requests from riders seeking transportation services and related collection activities using our Lyft platform. As a result, our single performance obligation in the transaction is to connect drivers with riders to facilitate the completion of a successful transportation service for riders.
> Because end-users access our platform for free and we have no performance obligation to end-users, end-users are not our customers.
That's some bracing cynicism there. So their business plan is to keep paying their taxi drivers less (with VC money), or providing worse service, until they figure out self-driving cars or run out of money?
It's actually the opposite. You're misunderstanding "end-users" and projecting your inherent bias against Uber. Once I explain what it _actually_ means, you'll probably project it again but here you're just doing so incorrectly :)
"End-users" as defined here are actually the riders/"eaters" of Uber. Uber is saying these people are not their customers because they are just providing access to drivers/restaurants+delivery people with their platform. What Uber is saying is that driver/restaurants/devliery people are independently selling their products/services on Uber as a platform and that end-users (riders/eaters/consumers) are simply given free access.
Calling it now, Uber has several large loans coming due in 2021 and 2022. I suspect we will see the bankruptcy proceedings around that time.
They may survive, not strapped with the debt, perhaps they’ll be profitable. Aka they will just have debt restructuring. Regardless, that’s when I suspect the bankruptcy will occur.
How is Uber, a company that has never made money, and has competitors in every market in taxi cabs, too big to fail? TBTF refers to banks, not side investments from SoftBank and Saudi Arabia.
It's hard to see how they're 'too big to fail'. That term is used to describe institutions whose demise could wreak havoc on society, so their failure is prevented by government.
Uber could fail and no-one but its investors would bat an eyelid.
From the drivers I know, they much prefer to drive for Lyft. Almost entirely because they say Lyft riders leave much larger tips and Uber drivers tip much less if at all.
When the business matures, the margins in the TNC business will be razor thin.
There will be local competitors in every city. There will be pure software companies that build TNC apps and sell operating services. Maybe Vuitton, Apple, Waymo, Herz, Hilton, Star Alliance, Oneworld etc. also have their own mobility service provider networks.
Say 4 billion people, 1% of them do ride-hailing of some kind, every one of them has five ride sharing apps and they pay $0.20 per day for each. That's $15B in revenue total for all those involved.
They both got the stock ticker that matches the company name, but could that be a bad thing? eg. If you google "lyft" it brings up the company and an Ad to drive for lyft, so to see the stock price (in G) you have to type "nasdaq:lyft".
While Uber looks healthier especially given it's size, Lyft's reception after the IPO was quite negative, with the stock falling 30% immediately in the days that followed.
Even with better economics and larger scale the larger question looms, which is how do these businesses turn profitable.
Are they subsidizing a market and creating it by charging less, or will they reach a certain scale and be able to cut back on certain expenses which will give them profitability.
I don't think Amazon is a fair comparison here, because they were investing in infrastructure, building out global logistics, which is different from Uber and Lyft. They should be much more profitable because they don't have to invest in that. They don't know warehouses and logistics, and everything else that e-commerce required.
Also in the case of Amazon the supposed profits would be reinvested in the business with large CapEx spend, here it seems that the negative margins are being spent on sales and marketing, which isn't the same as what Amazon was doing when it was treading in the negative for a decade.