The Value of Deferred Money in the Chris Davis Deal

On Saturday, the Orioles agreed to re-sign star first baseman Chris Davis, and the price tag was staggering given his limited market: $161 million over seven years. Except, they aren’t really paying him $161 million over the next seven years; as part of the deal, the Orioles have deferred about a quarter of the contract, pushing $6 million per year of his annual salary into the years after the contract has expired, and pushing some of it way into the future. Under the terms of this deal, Baltimore will now be sending Davis checks for the next 22 years, with the $42 million in deferred money being paid out from 2023 through 2037.

So while the $161 million number is a headline-grabber, the actual value of this deal is quite a bit lower than that, since money loses value over time. We talked about this last year with the Max Scherzer deal, and not surprisingly, Davis and Scherzer are both represented by Scott Boras; he’s clearly willing to create long-term payment structures in order to get a larger total payout numbers when the deals are announced. And for the owners, these types of deals help them acquire (or retain) star players while potentially pushing the future costs of the deal onto another ownership group; Peter Angelos is 86 years old, and will probably not be alive for the entire duration of Davis’ payments.

But to compare these deferred-money contracts to deals signed where the money is paid out during the time the player is actually under contract, we need to use some accounting techniques to create apples-to-apples comparisons. Using net present value (NPV) calculations, we can measure just how much money the Orioles saved by deferring $42 million of Davis’ contract, and see what an equivalent contract without deferrals would have been.

In last year’s post on Scherzer, I used a 7% discount rate, but that was too aggressive on my part; the league and the MLBPA have agreed to use a 4% discount rate for calculating the present value of long-term deals. Since that’s what they’ve agreed upon, and they have better access to future cash flow numbers than we do, it’s probably best for us to accept their discount rate calculations than argue for some other number. So, using that 4% number, we can calculate not only the net present value of Davis’ deal, but also the equivalent value of different payment structures.

Chris Davis Contract Structures
Year Deferred Flat Backloaded EqualNPV
2016 $17,000,000 $23,000,000 $17,000,000 $21,127,300
2017 $17,000,000 $23,000,000 $17,000,000 $21,127,300
2018 $17,000,000 $23,000,000 $22,000,000 $21,127,300
2019 $17,000,000 $23,000,000 $24,000,000 $21,127,300
2020 $17,000,000 $23,000,000 $25,000,000 $21,127,300
2021 $17,000,000 $23,000,000 $28,000,000 $21,127,300
2022 $17,000,000 $23,000,000 $28,000,000 $21,127,300
2023 $3,500,000
2024 $3,500,000
2025 $3,500,000
2026 $3,500,000
2027 $3,500,000
2028 $3,500,000
2029 $3,500,000
2030 $3,500,000
2031 $3,500,000
2032 $3,500,000
2033 $1,400,000
2034 $1,400,000
2035 $1,400,000
2036 $1,400,000
2037 $1,400,000
Total $161,000,000 $161,000,000 $161,000,000 $147,891,100
NPV $126,807,204 $138,047,257 $136,091,513 $126,807,210

Because of the deferred money in the deal, the NPV of Davis’ contract at a 4% discount rate is $126.8 million; that’s the equivalent lump-sum payment that would have the same value as the payment structure he accepted. And this number lines up almost exactly with the number Buster Olney was given by a management source about the value of the deal from their perspective.

Clearly, $127 million is a lot lower than $161 million, and if we just stopped there, it’d be much easier to defend this outlay on the Orioles part. After all, $127 million is almost exactly what I predicted Davis would get back before the off-season began, and that was on a five-year contract, not a seven-year deal. But we don’t want to just take the NPV of this deal and compare it to other total contract numbers, since teams aren’t making lump-sum payments on those deals either. The numbers that really matter are the ones in the other columns; the NPVs of the alternate payout structures, and the total dollars that the team would have had to pay to come up with an equal NPV to this deal.

As you can see from looking at those columns, the NPV of a seven year, $161 million contract with a flat payout structure, paying $23 million per season from year one through year seven, is $138 million. In other words, the Orioles saved about $11 million in long-term value by deferring the $42 million well down the road. The savings is more like $9 million if we think they could have heavily backloaded the deal instead, starting the contract at $17 million but ending up at $28 million in the final two years of the contract; the NPV on that kind of structure would be $136 million.

But the fourth column is probably the one we care about the most. Generally, we care about contract values because we’re using them to answer the question of what else the team could have done, and we want to know the currency value of this contract; what they could have spent on someone else if they wanted to use this money to sign another player, perhaps one less open to deferrals. So, by fixing the NPV at that $126.8 million figure, we can come up with an equivalent total contract number. Using that 4% discount rate on a seven-year deal, a $147.8 million contract with a flat payout structure would have the same value as the contract Davis just signed.

So, realistically, the deferred money in this deal makes this contract equal to a seven year, $148 million deal that was paid out over the time that Davis is required to play for the team. That’s the real value of this deal when comparing it to other contracts signed this winter. Of course, the fact that he didn’t get an opt-out also has to be factored in as a positive for the Orioles, but he did get a no-trade clause, which has some additional value to Davis. No contract comparison is as simple as what we did above, but when looking at the total dollar figures being bandied about, it’s better to think of Davis’ contract as a $148 million deal. That’s the non-deferred equivalent.



Print This Post



Dave is the Managing Editor of FanGraphs.


Sort by:   newest | oldest | most voted
tz
Member
tz
4 months 8 days ago

4% seems about right. If you wanted to buy an annuity for long-term payments from folks that specialize in that kind of stuff, the price will reflect about a 3.5% to 4% discount rate.

phillyfan
Member
phillyfan
4 months 7 days ago

Here’s another way of looking at it from Davis’s perspective. I went to a website for structured settlements and entered the numbers for the $42MM in deferred payments into a calculator. If CD wanted to cash out that $42MM portion of his contract today, a structured settlement firm would pay him in the range of about $14MM to $18MM. That suggests he signed for something between $133MM and $137.5MM

ItsLudovic
Member
ItsLudovic
4 months 7 days ago

Presumably there’s a significant difference between what Davis’s $42 million is actually worth and what a structural settlement firm would pay him for it.

bluejaysstatsgeek
Member
4 months 8 days ago

To any of my students reading this, Dave anticipated a question on your finance midterm!

Mooser
Member
Mooser
4 months 8 days ago

So based on the $8M/win theory, O’s are projecting around 3.8 WAR in 2016. Not that crazy.

Age WAR $/Win
30 3.8 $8,000 $30,400
31 3.3 $8,400 $27,720
32 2.8 $8,820 $24,696
33 2.3 $9,261 $21,300
34 1.8 $9,724 $17,503
35 1.3 $10,210 $13,273
36 0.8 $10,721 $8,577
$143,470

Phillies' Front Office
Member
Phillies' Front Office
4 months 8 days ago

When we say, “wins cost $8MM per”, we’re assuming teams buy all their wins in the contract at that price, including future wins. It’s a simplistic rule of thumb, but it’s a simplistic rule of thumb that’s baked into the original assumption.

Davis projects for 3.2 wins this year, the depth charts give him 3.3 Ignoring the fact that his comps aren’t wonderful (http://www.fangraphs.com/blogs/chris-davis-and-the-free-agent-bottleneck/), his type (high K%, ISO) are aging worse than we expected (http://www.fangraphs.com/blogs/component-changes-in-new-hitter-aging-curves/), and applying the typical -.5 WAR assumption gives:

30 3.3
31 2.8
32 2.3
33 1.8
34 1.3
35 .8
36 .3
Total: 12.6 wins, about $100M exactly. To me, this deal reeks of paying for past value. You can apply more generous assumptions and fiddle around the edges if you want, but even that $8 M per win assumption isn’t at NPV, it’s at sticker price. By that metric, you’d have to finagle a way to, as Cameron pointed out, about $148 MM in expected value. I don’t know how the Orioles are getting themselves there.

NickChristy
Member
NickChristy
4 months 8 days ago

Davis projects for 3.2 wins this year by Fangraphs*

How the Orioles arrived at their valuation probably has absolutely nothing to do with fangraphs evaluation. I doubt their head of analytics uses these projections, or they wouldn’t be projected to be a under .500 team. Each team has it’s own database, projections, etc.

Phillies' Front Office
Member
Phillies' Front Office
4 months 8 days ago

I’m sure they’ve got a proprietary projection system. I doubt it’s much better than any of the other 29 proprietary projection systems in baseball, or for that matter, that it’s dramatically better than steamer or ZiPS.

Either way, it would only justify this contract if the Orioles honestly believe Davis to be a true talent, consistent 4-5 win player this year, and go from there. That’s an aggressive assumption, and one that’s not helped by the fact that the Orioles were Davis’ most prominent suitor, and the only one in the $150+ M universe.

Maybe their scouts see something they like. Maybe they have some secret reason to believe Davis can defy aging curves for high strikeout sluggers, and has found the secret to consistent greatness that has eluded him for so long. Whatever it was that sold the O’s on this contract that wasn’t Scott Boras, there’s no reasonable doubt that they paid way more than they had to, and way more than public saber research suggests they should’ve.

NickChristy
Member
NickChristy
4 months 8 days ago

I’m not sure you read the article. You wouldn’t be using the $161M to measure the “value” of the contract, you would be using $143M which equates to roughly 3.5-4 WAR starting this year. It’s entirely possible that the Orioles view him as that. They don’t actually need scouts to see his is capable of that. The biggest problem isn’t if he can put up numbers to justify the contract, it is if the season he had two years ago shows up again.

2013 – 7 WAR
2014 – 0.8 WAR
2015 – 5.6 WAR

So – this is where projections are VERY difficult to use. You have to trust that your intel says that 2014 is a fluke and he is more like 2013/2015 version. Most likely this contract will be a bargain or a bust, but highly unlikely he is a 3-4 win player next year and follows a consistent path.

Paul22
Member
Paul22
4 months 8 days ago

The future wins will be more expensive, increasing by at least 5% per year. MLB revenues increases have been more like 10% per year, and payroll inflation about 8%, so 5% is conservative.

Phillies' Front Office
Member
Phillies' Front Office
4 months 8 days ago

I’m well aware. When you estimate the price of a win as $8 M, you don’t use the increasing cost of wins in the future in your calculation.

From here (http://www.fangraphs.com/blogs/the-cost-of-a-win-in-the-2014-off-season/):

“Keep in mind, though, that if you decide you like the 10% discount rate model, you can’t use it to argue that the Cano deal was under the market rate of $6 million per win, because that estimate came from applying no discount to future spending. Each model has their own “market rate”, and so here are the baselines for each of the three calculations.”

You’re applying a future discount on wins from the $8 M figure that already includes that. You could try to distinguish between wins bought for 2016 vs. wins bought for 2020, but you’d end up having to knock down the $8M per win figure to apply that inflation. It’s much easier to use the $8 M figure, acknowledging that it’s a rough estimate of teams buying wins, and apply NPV like the article above.

NashvilleSounds
Member
NashvilleSounds
4 months 8 days ago

In 2023, the money is also not likely to be coming out of Peter Angelos’s bank account. That had to have some appeal to him.

Brad Johnson
Member
Member
4 months 8 days ago

Theoretically, it would come out of the sale price of his team. In reality, it’s probably too small an amount to register. If you’re implying he’s going to die, then that’s another story.

NickChristy
Member
NickChristy
4 months 8 days ago

Yes that is what he is implying. Everyone dies, and Angelos is 86 years old. He won’t be selling it before he dies and his son stands to take over (already involved in front office.)

Brad Johnson
Member
Member
4 months 8 days ago

Worth noting, if Davis continues to live in Texas, then his deferrals won’t have state income tax. Also assuming Texas doesn’t change its laws between now and when they kick in. That’s worth about $2.1MM without regard to discount rates.

Paul22
Member
Paul22
4 months 8 days ago

He can live anywhere he wants that has a zero tax rate. There will always be at least 1 state to choose from. The saving may actually be higher if any of the states he plays in frequently raises their tax rates. The high rates in NY and Tor make the AL East one of the more expensive divisions tax wise, even with Tampa Bay at 0

bglick4
Member
bglick4
4 months 8 days ago

Yes, but does he pay this tax on the deferred money or will he only pay tax based on where he is living when it’s paid? I’m guessing the latter, but in a country where you’re “taxed” because you can’t afford health insurance, I guess anything’s possible.

calhoucr
Member
calhoucr
4 months 8 days ago

The source of the income for state tax purposes depends on how the payment is structured and the reach of Maryland’s tax laws when the payments are made. Federal law prohibits a state from imposing tax on a non-resident for any amounts that are paid through certain qualified accounts (e.g., a qualified retirement plan). But a state is free to impose tax on deferred compensation paid to a non-resident if the compensation was earned while performing services within the state (e.g., as a salaried employee of the Orioles) and the amounts are not paid through a federally qualified plan.

So if the deferred payments are made to Chris Davis as deferred wages, Maryland may be able to tax some percentage of them (whatever percentage is attributable to services performed in MD – probably less than 50%, given that spring training / other mandatory work days must be taken into account). Hypothetically, other states could also try to take a piece of the pie (e.g., New York or MA for road games), but that’s treading into very aggressive taxation territory and may violate due process.

But (a) Maryland may choose not to impose tax on such income, even though it is permitted to do so (I can’t find any authority on this point); and (b) Boras’s team may well have anticipated this issue and structured the payments so that they are either paid through a qualified plan (unlikely) or solely attributable to Texas or some other non-income-tax state.

Avi24
Member
Avi24
4 months 8 days ago

Does it benefit the Orioles as well if Davis continues to live in Texas or it only benefits Davis?

Les Vegetables
Member
Les Vegetables
4 months 8 days ago

Only thing I don’t get: I only get a $126.5 million NPV when I use 2016 as the first time period, but should the time period for 2016 be 0 since that is the year we’re in? Isn’t the NPV of $17 million in 2016 $17 million? Or are we assuming over the course of 2016 a 4% discount still applies?

Shauncore
Member
Shauncore
4 months 8 days ago

I think you may be correct. Here’s a chapter from Wharton at Penn (arguably the #1 finance grad school in the country) and it specifically says:

“Since the first cash flow occurs 0 years in the future, or today, it does not need to be adjusted.”

http://finance.wharton.upenn.edu/~acmack/Chapter%204%20Solutions%20V4.doc

jianadaren
Member
jianadaren
4 months 8 days ago

Depends if you’re assuming the payments come “in advance” (i.e. January 1st, 2016) or “not in advance” (i.e. December 31st, 2016).

It actually might be wise to split the difference, since salaries are actually paid during the baseball season, the middle of which is around July 1st.

Deke
Member
Member
Deke
4 months 8 days ago

Coming from a Finance background, the default assumption for the standard present value calculations assume that the payments come at the END of each period. So, when calculating annual NPV, the default would be to assume that his entire 2016 salary was paid at the end of the year.

For a more accurate calculation, you can break the annual payments into equal monthly amounts, divide the discount rate by 12 to find the monthly discount rate (.04/12 = .0033), then apply that to the monthly amounts that would be assumed to be paid at the end of the month.

With that said and without running the calcs myself, using annual numbers should be accurate enough for this purpose.

Shuki
Member
Shuki
4 months 7 days ago

You could also use a quick expedient, assuming that the payouts are being done on a straightline basis (i.e. equally spread-out throughout the year) and use a half year-convention for the discounting. For year-1 you discount using half a year, year-2 gets a 1.5 discount rate and so on and so forth.

Even though excel doesn’t allow for this directly in its NPV formula you can get to it through pretty quick calculation.

Cash Flow 1 – 100
Discount Rate 4%
Period – 1

Formula (Cash Flow)/(1+ discount rate)^(Period minus .5) = (100)/(1+.05)^.5

This approximates the assumption that half of the cash during the year has a shorter discount period and half has a longer discount.

Deke
Member
Member
Deke
4 months 8 days ago

This is a great article, and how ALL salary discussions should be handled in the mainstream media. Contract numbers for various players are thrown around and compared with no accounting for the time value of money, so it’s like comparing apples to pineapple guavas in many situations.

I’m having a tough time seeing how a major league team should be using a 4% discount rate. Typically, an enterprise would use their Weighted Average Cost of Capital (WACC) as their discount rate, which is the weighted costs (% per year) of the various sources of capital the enterprise obtains from it’s creditors and owners to fund it’s business. A 4% cost of capital on debt financing MAY be reasonable based on the how well a baseball team cash flows and covers their potential debt, but I would assume that any baseball owner would expect WELL above a 4% rate of return on their invested capital. Something closer to Dave’s 7% rate used previously would make more sense if equity demands returns in the teens and the company can borrow at a VERY low rate. I would continue to go with that rate, or slightly higher, for accuracy regardless of what the league and MLBPA have agreed to . . . It would also be interesting to try and estimate different teams WACC based on what is known about their capital structures (debt vs. equity) and the likely costs of each, as this could help better understand the different contract structures and amounts handed out by teams . . . for the bucket list!

This also misses the distinction between the team’s and the player’s perspective. Players will most definitely have a different discount rate than teams, and this difference should be accounted for when evaluating a deal from both sides (who did better in the deal?). A player would be coming from a perspective closer to an investor, namely using Opportunity Cost of Capital. If a player gets a dollar today vs. at a later date, that dollar could be invested at a certain expected rate of return elsewhere in the meantime, and the opportunity cost of capital is that rate that could have otherwise been earned. The rate will differ depending on the market (current “risk-free” rate and how much the market compensates for risk above and beyond “risk-free”) and the player’s tolerance for taking risk (potentially losing money). Tax implications and other factors should also be accounted for in the cash flows as previously noted by another commenter.

Again, with all that said Dave does a great job of explaining and using these concepts in evaluating player contracts, and I wish all of the other media outlets would do the same. Either way, it’s important us baseball nerds understand these concepts in detail and use them appropriately when analyzing contracts (or any baseball financial decision) for those analyses to be accurate.

bc231
Member
bc231
4 months 8 days ago

This signing also cost the O’s a sandwich round draft pick, which is probably worth another ~10 million in NPV. Not enormous, but it pushes the cost ever higher.

Shirtless Bartolo Colon
Member
4 months 8 days ago

My sandwich round picks:

5. Pepperidge Farms
4. Arnold
3. Bimbo toasted bread
2. Nature’s Own
1. Two deep-fried grill cheeses cut in circles.

waldthm
Member
waldthm
4 months 8 days ago

NPV is confusingly applied in baseball. And it should be different depending on if you are the player versus club.

For players, unless they do something criminal, the contract payments are essentially riskless (no club to my knowledge has ever restructured player contracts due to bankruptcy filings). If players are indifferent between the club versus the U.S. government paying the same salary (which is considered to be riskless), Chris Davis should use a discount rate equal to the 20-year U.S. Treasury yield, or 2.6%. The PV of the contract should then be $137 million.

For management, it’s more complicated given the risk involved in the expected cash inflows from the player. For the expected cash inflows, if you assume $8 million / WAR in 2015, and WAR inflation of 2% per annum (historical average), and use historical comps to project future WAR for Chris Davis through 2022, the IRR would be 26%. The floor of club’s hurdle rate is thus 26% (higher is better given cash outflows come later).

This could be consistent with the risk involved in such a contract. However, it’s never great to apply capital budgeting tools on baseball contracts, as it’s difficult to calculate the cash flows correctly (e.g., Davis’s presence could positively affect WAR of teammates, future merchandising revenues, 1 WAR = $8 million cash is bad model, comps are bad, etc.). I would love to see how Friedman’s front office calculates break-even for players.

Johnston
Member
Member
Johnston
4 months 8 days ago

Davis has such an erratic record – go pull up BBREF and look at his WAR over his career – that a long-term contract looks extremely risky. We may very well be looking at the next Ryan Howard contract.

IHateJoeBuck
Member
IHateJoeBuck
4 months 8 days ago

Or pull it up on another website that could show his career WAR. Perhaps that website could also provide graphs for fans.

Dillon M
Member
Dillon M
4 months 7 days ago

The truth is MLB teams should discount all player contracts given out to present value but only for internal usage purposes. While the present values of contracts are useful for teams and the player’s association, they are merely just tidbits of fun facts for outside parties. A smart team will discount the salaries to present day but will also discount the total dollar values of the player’s projected WAR to present day. The present value of the player’s WAR (which will be the “cash” inflow) minus the present value of the player’s salary (clearly the cash outflow) will give you the net present value of the contract. So while it’s cool to know the present value of a contract, the market rate should probably be taken at its face value. It kind of relates to the differences between financial (external reporting) and managerial/cost (internal reporting for decisions) accounting.

wpDiscuz