Composition and safety of TIAA traditional annuity

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
User avatar
Gabriel A. Lozada
Posts: 37
Joined: Wed Oct 16, 2019 10:53 pm
Location: Dept. of Economics, University of Utah, Salt Lake City, Utah, USA 84112
Contact:

Re: Composition and safety of TIAA traditional annuity

Post by Gabriel A. Lozada » Thu Oct 24, 2019 4:58 pm

Nittany_Lion wrote:
Thu Oct 24, 2019 10:08 am
[...] perhaps the best course of action [...]
Thank you, Nittany_Lion, for your thoughtful opinion on what to do going forward. Historical statistics and modeling such as I did can inform such a discussion, but they can only do so much; the future always has an element of mystery, sometimes large, sometimes smaller.
Nittany_Lion wrote:
Thu Oct 24, 2019 10:08 am
[...] to view Traditional, circa 2019, as being akin to a barely investment grade corporate bond fund [...]
This made me think of another, perhaps minor, point. The TIAA Traditional white paper mentioned in my paper (for the current version of the white paper see https://www.tiaa.org/public/pdf/complia ... -paper.pdf) points out: "TIAA’s high ratings do not mean that its General Account investments are restricted to securities that carry equally high ratings." One reason (in my opinion) is diversification. Diversification is available to mutual fund/ETF investors as well: if an investor's bond holdings are well-diversified, a bond-only investor should think of himself or herself as having a higher rating than the average rating of the bonds they are invested in. (I am not here giving advice to lower the quality of anyone's bonds: there is a reason insurance regulators prohibit insurance companies from large investments in low-quality bonds. Also, most individual investors hold stocks as well as bonds, and that may have a large effect on the type of bonds one should hold.) Another reason for TIAA's high rating, as described by the Moody's rating agency report available at https://www.tiaa.org/public/about-tiaa/ ... nt-library, is that "TIAA has an outstanding liquidity position, as evidenced by it having one of the strongest results in our liquidity stress testing for US life insurers. TIAA had approximately $212 billion in general account liabilities as of December 31, 2018. Only approximately 28% of that amount is withdrawable upon demand by participants." An individual mutual fund investor could only mimic this advantage of TIAA by refraining from liquidating his or her bond investments except over a long period of time. That in turn would require the investor to have unshakable confidence in the bond investments he or she holds.

User avatar
Gabriel A. Lozada
Posts: 37
Joined: Wed Oct 16, 2019 10:53 pm
Location: Dept. of Economics, University of Utah, Salt Lake City, Utah, USA 84112
Contact:

Re: Composition and safety of TIAA traditional annuity

Post by Gabriel A. Lozada » Thu Oct 24, 2019 5:22 pm

aristotelian wrote:
Thu Oct 24, 2019 1:53 pm
Apologies if this is simplistic. I just skimmed the paper and most of it is over my head. However, wouldn't we expect TIAA Traditional to underperform the holdings of the General Account? The General Account has to pay for TIAA Traditional plus at minimum some expenses for TIAA to keep the program going. Of course individual investors would be better off replicating the General Account if they could. Am I missing something?
I think what you're missing is:
Gabriel A. Lozada wrote:
Sun Oct 20, 2019 11:45 pm
[...] The performance data I use in the paper is for TIAA Traditional, not for the General Account; in Section 4 it comes from Goodman and Richardson (2014), supplemented with sometimes-contradictory more recent data from TIAA, and in Section 3 it comes from Babbel et al. (2015). [...]
Page 14 of my paper describes how I used the data from Babbel et al. (2015); the Appendix describes how I used the more recent data from TIAA; and page 19 describes how I used the data from Goodman and Richardson (2014).
Gabriel A. Lozada wrote:
Sun Oct 20, 2019 11:45 pm
The role of the General Account in my paper is to determine which combination of Vanguard funds should be used to benchmark Traditional [...]
That is, the paper uses the General Account's quality and duration, but not its performance. I actually don't have data on the General Account's performance, although one might be able to infer it from reports from the rating agencies.

Thank you for your interest in the paper!

User avatar
Gabriel A. Lozada
Posts: 37
Joined: Wed Oct 16, 2019 10:53 pm
Location: Dept. of Economics, University of Utah, Salt Lake City, Utah, USA 84112
Contact:

Re: Composition and safety of TIAA traditional annuity

Post by Gabriel A. Lozada » Thu Oct 24, 2019 9:38 pm

TSR wrote:
Thu Oct 24, 2019 2:53 pm
What a fascinating thread representing the best of what Bogleheads has to offer. I'm chiming in to say that when my girlfriend was starting her academic job several years ago, she could choose between TIAA and Fidelity, and she was advised by all of her older colleagues to go with TIAA traditional rather than the Fidelity option. They gave her a "just trust me, it's been good to us" kind of explanation. She and I had been steeping ourselves in Bogleheads advice and so I argued that "TIAA can't possibly be that much better than what bond funds can usually offer, so if you can stay the course on bond funds then why not just go with the less opaque investment vehicle?"

My takeaway from this article is that I was right; also, her colleagues were right. That is, this is the rare case where the two options are both pretty good and it's hard to make a strong argument either way. My other takeaway is that there are probably a lot of senior-level professors out there saying the same thing to more junior ones, and there ought to be some better advice than "just trust me." Perhaps that advice is something like, "If you don't want to spend a lot of time thinking about your investments, stick with TIAA for your fixed income. But if you know what you're doing or already have a portfolio with a fixed-income component, you might want to stick with one of Fidelity's low-cost bond funds."

Thanks for the very interesting discussion.
Thank you for your interesting overview.

Since you mentioned Fidelity's low-cost bond funds, I was wondering how to construct a "duration- and quality-mimicking" portfolio, as in the paper's Figure 1, using low-cost Fidelity funds. [Most readers have never wondered this and won't be interested in the rest of this post.]

It'd be straightforward enough to match duration using Fidelity's Intermediate Treasury Bond Index Fund and Fidelity's Long-Term Treasury Bond Index Fund, but I was not able to find a less-than-BBB-quality Fidelity fund to blend with those in order to match the General Account's quality while restricting the expense ratio to below 0.4. Also, Fidelity's GNMA fund has an expense ratio of 0.45, so matching "bond types" would not be possible with less-than-0.4-expense-ratio Fidelity funds. (The Fidelity Mortgage Securities Fund has the same 0.45% expense ratio as the Fidelity GNMA Fund.)

If one relaxes my "0.4 expense ratio" constraint then there is more than one Fidelity portfolio that mimics the General Account. Given that both of the Fidelity Treasury Bond Index funds I mentioned above have very low expense ratios (0.03%), while the Vanguard GNMA fund the paper uses charges 0.21% (Admiral share class 0.11%), the Vanguard High-Yield Corporate charges 0.23% (Admiral share class 0.13%), and the Vanguard Long-term Investment-Grade Fund charges 0.22% (Admiral share class 0.12%), it's natural to ask whether one can find a mimicking Fidelity portfolio whose overall expense ratio is less than that of some of the paper's mimicking Vanguard portfolios. The answer is yes, there is at least one: 78% Fidelity Intermediate Treasury Index, 13% Fidelity Long Treasury Index, and 9% Fidelity High Income Fund. This mimics the General Account and has a weighted expense ratio of 0.09%. (I used durations of 6.38, 18.30, and 2.63, and a "1 minus loss ratio" of 0.8594 for Fidelity High Income.) However, its bond types are 91% government, 0% mortgage, and 9% other, giving a "bond type dist." (as in row 9 of the paper's Table 3) of 62%, which is much higher than the Vanguard alternatives of Tables 3 and 4 because they have more high-quality corporate bonds (and several have more mortgages).

Another Fidelity duration- and quality-mimicking portfolio, Fidelity GNMA 58%, Fidelity Long Treasury Index 33%, and Fidelity High Income 9%, has a better, but still not good, "bond type dist." of 51%, and a weighted expense ratio of 0.33%, which is higher than Vanguard's. (I used a duration of 2.32 for Fidelity GNMA.)

It is hard to construct a duration- and quality-mimicking portfolio using Fidelity funds which has a good "bond type dist." because Fidelity only has one long-term taxable bond fund and it only contains Treasuries. Together with trying to match the General Account's 34% holdings of mortgages, it is hard to get a sufficiently long duration and get close to the General Account's holdings of 50% corporate bonds. The best I could come up with in the limited amount of time I spent thinking about it is a four-fund portfolio using 47% Fidelity Corporate Bond Fund (expense ratio 0.45%), Fidelity GNMA 29%, Fidelity Long Treasury Index 18%, and Fidelity High Income 6%. This gives a "bond type dist." of 9%, which is better than the Vanguard 2- and 3-fund portfolios but not nearly as good as the Vanguard 4-fund portfolios, and has a weighted expense ratio of 0.39%. For the Fidelity Corporate Bond Fund I used a duration of 7.37 and a "1 minus loss ratio" of 0.9910.

A one-fund portfolio of the Fidelity Corporate Bond Fund comes close to mimicking the General Account's 7.59 duration and "1 minus loss ratio" of 0.9874, but again, there is that 0.45% expense ratio, and it only has one type of bond.

dknightd
Posts: 2020
Joined: Wed Mar 07, 2018 11:57 am

Re: Composition and safety of TIAA traditional annuity

Post by dknightd » Fri Oct 25, 2019 8:09 am

Gabriel A. Lozada wrote:
Thu Oct 24, 2019 9:38 pm


Since you mentioned Fidelity's low-cost bond funds, I was wondering how to construct a "duration- and quality-mimicking" portfolio, as in the paper's Figure 1, using low-cost Fidelity funds. [Most readers have never wondered this and won't be interested in the rest of this post.
I wonder if it even makes sense to try and replicate/mimic the TIAA general fund.
I think we have determined that the obligations of the TIAA fund may not match the obligations of a specific investor.
We have also seen that the composition of the general fund changes with time.
It is an interesting academic exercise, but would it make sense for the individual investor to do this? Perhaps, perhaps not. I guess it depends on the investor. . .

edit to close the italics
If you value a bird in the hand, pay off the loan. If you are willing to risk getting two birds (or none) from the market, invest the funds.

TSR
Posts: 855
Joined: Thu Apr 19, 2012 9:08 am

Re: Composition and safety of TIAA traditional annuity

Post by TSR » Fri Oct 25, 2019 10:03 am

Gabriel A. Lozada wrote:
Thu Oct 24, 2019 9:38 pm
Thank you for your interesting overview.

Since you mentioned Fidelity's low-cost bond funds, I was wondering how to construct a "duration- and quality-mimicking" portfolio, as in the paper's Figure 1, using low-cost Fidelity funds. [Most readers have never wondered this and won't be interested in the rest of this post.]

Fascinating, and I appreciate the analysis. After reading this, I feel pretty good in my suggestion to her that she just stick with Fidelity's total bond fund (or intermediate, or whatever they call it) for her fixed income component. :D

Topic Author
Nittany_Lion
Posts: 31
Joined: Fri Jul 20, 2018 10:49 am

Re: Composition and safety of TIAA traditional annuity

Post by Nittany_Lion » Fri Oct 25, 2019 2:20 pm

Just to be clear, I was only ruminating on my own personal situation; not suggesting a plan for similarly situated investors (although my wording may have made it sound that way).

I computed the current SEC yield on Gabriel's 4mt portfolio, which I believe was the best overall match to Traditional in terms of duration, quality, and bond type. By my calculations, that portfolio would have a yield of approximately 2.72% using investor-class shares. If you believe TIAA's costs are lower than the fees on these funds, then perhaps tack on another 10 basis points? That puts the mutual-fund version of Traditional within striking distance of the 3% guarantee.

I remain concerned about the nontrivial deterioration in credit quality since 2015, along with the slight increase in duration, both of which suggest the account's managers may be "reaching for yield" as they say.

More particularly, I worry how the account would hold up in a 2008-style crisis with its now riskier investments. Moreover, the 10-year Treasury rate was 3.74% at the beginning of 2008, compared to 1.77% today.

All of that said, the account does benefit from substantial reserves and stringent limits on the timing of its liabilities, as Gabriel pointed out. I also recall TIAA noting somewhere that money tends to flow into the fund during times of financial stress, which would reduce their need to sell assets during a meltdown.

How this all nets out in terms of the viability of the guarantee at low rates is beyond my ability to discern. Perhaps the upshot is that Traditional is not a free lunch, even at low rates, because of uncertainties about the guarantee and the chaos that would ensue if they failed to meet it. Put differently, when matched mutual fund yields are below 3%, TIAA is either the best thing going or a ticking time bomb, and it seems impossible to determine when it will transition from the former into the latter. If anyone figures that out, let me know!

User avatar
Gabriel A. Lozada
Posts: 37
Joined: Wed Oct 16, 2019 10:53 pm
Location: Dept. of Economics, University of Utah, Salt Lake City, Utah, USA 84112
Contact:

Re: Composition and safety of TIAA traditional annuity

Post by Gabriel A. Lozada » Sat Oct 26, 2019 3:44 pm

dknightd wrote:
Fri Oct 25, 2019 8:09 am
Gabriel A. Lozada wrote:
Thu Oct 24, 2019 9:38 pm
Since you mentioned Fidelity's low-cost bond funds, I was wondering how to construct a "duration- and quality-mimicking" portfolio, as in the paper's Figure 1, using low-cost Fidelity funds. [Most readers have never wondered this and won't be interested in the rest of this post.]
[...] It is an interesting academic exercise, but would it make sense for the individual investor to do this? Perhaps, perhaps not. I guess it depends on the investor. . .
I agree; it depends.

User avatar
Gabriel A. Lozada
Posts: 37
Joined: Wed Oct 16, 2019 10:53 pm
Location: Dept. of Economics, University of Utah, Salt Lake City, Utah, USA 84112
Contact:

Re: Composition and safety of TIAA traditional annuity

Post by Gabriel A. Lozada » Sat Oct 26, 2019 4:30 pm

Nittany_Lion wrote:
Fri Oct 25, 2019 2:20 pm
Put differently, when matched mutual fund yields are below 3%, TIAA is either the best thing going or a ticking time bomb, and it seems impossible to determine when it will transition from the former into the latter. If anyone figures that out, let me know!
Perhaps you will want to keep tabs on the reports of the rating agencies, which TIAA is to be commended for placing on their web site: https://www.tiaa.org/public/about-tiaa/ ... nt-library. Fitch's report "does not view [the 3% minimum guaranteed rate] as a near-term issue." However, rating agency reports are not perfect; for example, S&P's is written using "Macroeconomic Assumptions" of a 2019 ten-year Treasury rate of 3.3% (it is now 1.8%) and a 2019 AAA corporate rate of 4.3% (it is now 2.3% for 10-year AAA corporates).

Topic Author
Nittany_Lion
Posts: 31
Joined: Fri Jul 20, 2018 10:49 am

Re: Composition and safety of TIAA traditional annuity

Post by Nittany_Lion » Tue Oct 29, 2019 2:46 pm

I will definitely check them out. Thanks, Gabriel.

George Kaplan
Posts: 5
Joined: Wed Feb 12, 2020 11:14 am

Re: Composition and safety of TIAA traditional annuity

Post by George Kaplan » Wed Feb 12, 2020 11:33 am

I read the White Paper carefully. If "TIAA has concluded that contract value approximates fair value" then they believe that the duration of their liabilities is around zero. If the duration of TIAA's assets is about 8 years, then they must be exposed to gigantic interest rate risk.

Further, TIAA believes that the "10-year risk-free rate seems appropriate to use as a benchmark for the TIAA Traditional Annuity". The 10-year risk-free rate is a new money rate. This contradicts TIAA's non-profit mandate to "pass through" investment PORTFOLIO earnings to participants. The 89% correlation would totally disappear if rates turned up.

talzara
Posts: 896
Joined: Thu Feb 12, 2009 7:40 pm

Re: Composition and safety of TIAA traditional annuity

Post by talzara » Wed Feb 12, 2020 4:24 pm

George Kaplan wrote:
Wed Feb 12, 2020 11:33 am
Further, TIAA believes that the "10-year risk-free rate seems appropriate to use as a benchmark for the TIAA Traditional Annuity". The 10-year risk-free rate is a new money rate. This contradicts TIAA's non-profit mandate to "pass through" investment PORTFOLIO earnings to participants. The 89% correlation would totally disappear if rates turned up.
TIAA is a New York stock insurance company. The mandate of an insurance company is to pay claims without running out of reserves.

TIAA does not have to pass through the portfolio earnings to participants. It can choose to pay out part of the portfolio earnings, all of it, or even supplement it from reserves. As long as it pays the annuities as required by the contract, it is doing what an insurance company is supposed to do.

George Kaplan
Posts: 5
Joined: Wed Feb 12, 2020 11:14 am

Re: Composition and safety of TIAA traditional annuity

Post by George Kaplan » Thu Feb 13, 2020 9:40 am

TIAA has no external capital. Its reserves are built by "setting aside" investment earnings which it commits to later return to policyholders.

User avatar
Gabriel A. Lozada
Posts: 37
Joined: Wed Oct 16, 2019 10:53 pm
Location: Dept. of Economics, University of Utah, Salt Lake City, Utah, USA 84112
Contact:

Re: Composition and safety of TIAA traditional annuity

Post by Gabriel A. Lozada » Thu Feb 13, 2020 12:40 pm

George Kaplan wrote:
Wed Feb 12, 2020 11:33 am
I read the White Paper carefully. If "TIAA has concluded that contract value approximates fair value" then they believe that the duration of their liabilities is around zero. [...]
This would only be correct if the contract was subject to immediate liquidation upon demand of the contract holder. Many of TIAA Traditional's contracts are not at all subject to such liquidation. For those contracts, the "contract value" is the value of a withdrawal flow over nine years (starting when the policyholder requests withdrawal). The duration of such a liability is much greater than zero. The duration of TIAA Traditional's liabilities to policyholders of these contracts who are not requesting withdrawal is even longer.

George Kaplan
Posts: 5
Joined: Wed Feb 12, 2020 11:14 am

Re: Composition and safety of TIAA traditional annuity

Post by George Kaplan » Thu Feb 13, 2020 3:53 pm

Look at page 6 of the white paper and see how they define "contract value". It is the accumulated value.

On any given day, if interest rates spike 1% TIAA's bond values will drop 8%ish. But the "contract value" does not budge.

User avatar
Gabriel A. Lozada
Posts: 37
Joined: Wed Oct 16, 2019 10:53 pm
Location: Dept. of Economics, University of Utah, Salt Lake City, Utah, USA 84112
Contact:

Re: Composition and safety of TIAA traditional annuity

Post by Gabriel A. Lozada » Thu Feb 13, 2020 6:25 pm

George Kaplan wrote:
Thu Feb 13, 2020 3:53 pm
Look at page 6 of the white paper and see how they define "contract value". It is the accumulated value.

On any given day, if interest rates spike 1% TIAA's bond values will drop 8%ish. But the "contract value" does not budge.
Yes; TIAA Traditional uses book value accounting, not mark-to-market accounting. But with withdrawals limited to a gradual flow over nine years, book value accounting is more relevant than mark-to-market accounting. Indeed, for high-grade fixed-income investments, paying much attention to mark-to-market accounting is an error if it is impossible to quickly sell the investment, or even if it is possible to quickly sell the asset but the investor knows he or she will never wish to do so.

If I buy an investment-grade bond promising to pay $1000 five years from now, for example to cover a $1000 expense I anticipate incurring five years from now, then the daily, mark-to-market fluctuations of this bond are completely irrelevant to me: they are meaningless noise, and paying attention to them could only serve to confuse the situation. On the other hand if I've bought that bond betting that interest rates will fall within the next week, then its daily, or even minute-by-minute, mark-to-market fluctuations are quite important to me.

George Kaplan
Posts: 5
Joined: Wed Feb 12, 2020 11:14 am

Re: Composition and safety of TIAA traditional annuity

Post by George Kaplan » Fri Feb 14, 2020 10:30 am

Except that TIAA is making representations in the white paper about the fair value of its liabilities. (I suspect this is partly motivated by considerations of IFRS 13). TIAA evidently believes that the fair value of the nine-year stream is identical whether the 10-year rate is 2% or 4% or even -2%. Hmm.

TIAA also contends in the white paper that "there is no principal market where such contracts are traded". In actuality, the principal market is the 403(b) market where TIAA accepts voluntary pension contributions every day.

Valuethinker
Posts: 39433
Joined: Fri May 11, 2007 11:07 am

Re: Composition and safety of TIAA traditional annuity

Post by Valuethinker » Fri Feb 14, 2020 10:42 am

Gabriel A. Lozada wrote:
Thu Oct 17, 2019 2:36 pm
Trurl Klapaucius wrote:
Thu Oct 17, 2019 12:16 pm
[...] I come up with a weighted average credit quality of 0.969, which corresponds to a loss rate of 1 – 0.969 = 0.031, which is between that of BBB (1.19%) and BB (6.90%).
It's hard to know what to do with the NR ("not rated") bonds. One perfectly reasonable strategy is the one Trurl Klapaucius takes, which is to ignore them. However, technically, then the 0.9690 sum of the "weighted credit quality" only represents the quality of 100% - 1.15% = 98.85% of the holdings, so the average credit quality of that 98.85% of the holdings is 0.9690/.9885 = 0.9803 (actually, 0.9802 with no intermediate roundings). That's still between BBB and BB, closer to BBB than 0.9690 was, but not as close as the year-end 2015 portfolio was. Only if one were to assume a 100% loss rate on NR bonds would the average credit quality be as low as 0.9690.

Such technical details aside, we've learned from Trurl Klapaucius is that TIAA's General Account quality is somewhat worse now than it was at year-end 2015.
Not Rated bonds are perhaps private placement bonds?

By accepting lower liquidity the investor can improve upon the yield (the costs for the issuer are lower).

I would conjecture that these are investment grade, and a rating of BBB or BBB+ would be the most appropriate (since BBB is the rating of 48% or so of all non-financial corporate bonds, probably use that rating to be conservative; or, alternatively, just assume that the ratings distribution mirrors the rest of the portfolio?).

Valuethinker
Posts: 39433
Joined: Fri May 11, 2007 11:07 am

Re: Composition and safety of TIAA traditional annuity

Post by Valuethinker » Fri Feb 14, 2020 10:45 am

Nittany_Lion wrote:
Fri Oct 25, 2019 2:20 pm
Just to be clear, I was only ruminating on my own personal situation; not suggesting a plan for similarly situated investors (although my wording may have made it sound that way).

I computed the current SEC yield on Gabriel's 4mt portfolio, which I believe was the best overall match to Traditional in terms of duration, quality, and bond type. By my calculations, that portfolio would have a yield of approximately 2.72% using investor-class shares. If you believe TIAA's costs are lower than the fees on these funds, then perhaps tack on another 10 basis points? That puts the mutual-fund version of Traditional within striking distance of the 3% guarantee.

I remain concerned about the nontrivial deterioration in credit quality since 2015, along with the slight increase in duration, both of which suggest the account's managers may be "reaching for yield" as they say.

More particularly, I worry how the account would hold up in a 2008-style crisis with its now riskier investments. Moreover, the 10-year Treasury rate was 3.74% at the beginning of 2008, compared to 1.77% today.

All of that said, the account does benefit from substantial reserves and stringent limits on the timing of its liabilities, as Gabriel pointed out. I also recall TIAA noting somewhere that money tends to flow into the fund during times of financial stress, which would reduce their need to sell assets during a meltdown.

How this all nets out in terms of the viability of the guarantee at low rates is beyond my ability to discern. Perhaps the upshot is that Traditional is not a free lunch, even at low rates, because of uncertainties about the guarantee and the chaos that would ensue if they failed to meet it. Put differently, when matched mutual fund yields are below 3%, TIAA is either the best thing going or a ticking time bomb, and it seems impossible to determine when it will transition from the former into the latter. If anyone figures that out, let me know!
When faced with complex decisions in finance where there are many unknowns, the utility maximising solution is often to "split the difference". To go 50/5 each way for the decision. That way you will never have been 100% wrong which is psychically more beneficial than being 100% right.

Oregano
Posts: 19
Joined: Fri Nov 22, 2019 9:30 pm

Re: Composition and safety of TIAA traditional annuity

Post by Oregano » Fri Feb 14, 2020 5:19 pm

The special sauce of TIAA Traditional is that it is over-collateralized (i.e., assets exceed liabilities.) That's it. That's why it is rated AAA and that's why it can pay above market rates.

Thus, I don't think credit risk is a large concern. I also don't think rising interest rates are a terribly large concern, either. The long-term viability of the product will be at risk if interest rates stay low or go lower (and my hunch is that is the case). This will become a problem because some of their products have 3% guarantees and the cash flows from even the overcollateralized portfolio won't support that (they might not already). Many vintages are paying much more than 3%. Due to the lock ups to withdraw money and the fact that they are always getting new inflows, TIAA has a lot of flexibility to manage this over time. They would address permanently low rates by cutting interest rates on existing vintages and freezing contributions to products that have 3% guarantees. I would guess they will be able to pull this off so that owners principal value is not at risk over time, but the substantial excess yields that some vintages are currently earning are probably not going to last much longer. Enjoy them while you can.

George Kaplan
Posts: 5
Joined: Wed Feb 12, 2020 11:14 am

Re: Composition and safety of TIAA traditional annuity

Post by George Kaplan » Sun Feb 16, 2020 1:42 am

Agreed. To my mind, the biggest risk to TIAA Trad is the contractual right that CREF people have to transfer into Trad. Say int rates fell to -2% (not out of the realm of possibility). CREF money in 3% min contracts (RA, SRA) could walk right in. I'm not sure TIAA has any protection against this.

User avatar
Gabriel A. Lozada
Posts: 37
Joined: Wed Oct 16, 2019 10:53 pm
Location: Dept. of Economics, University of Utah, Salt Lake City, Utah, USA 84112
Contact:

Re: Composition and safety of TIAA traditional annuity

Post by Gabriel A. Lozada » Sun Feb 16, 2020 2:17 am

George Kaplan wrote:
Fri Feb 14, 2020 10:30 am
Except that TIAA is making representations in the white paper about the fair value of its liabilities. (I suspect this is partly motivated by considerations of IFRS 13). TIAA evidently believes that the fair value of the nine-year stream is identical whether the 10-year rate is 2% or 4% or even -2%. Hmm.
It is true that from the viewpoint of mark-to-market accounting, book value accounting is weird. The fundamental question, though, is whether TIAA Traditional is exposed to, as you suggested earlier, "gigantic interest rate risk." I do not think that it is, at least to the extent that its policyholders can only withdraw funds over a period of many years.
George Kaplan wrote:
Fri Feb 14, 2020 10:30 am
TIAA also contends in the white paper that "there is no principal market where such contracts are traded". In actuality, the principal market is the 403(b) market where TIAA accepts voluntary pension contributions every day.
What TIAA means is simply that one can't sell one's TIAA Traditional assets immediately to a third party (again, for those contracts which prohibit immediate withdrawal). In other words, TIAA Traditional contracts are not at all like "bankbooks" during the Great Depression:
nisiprius wrote:
Mon Jul 09, 2018 12:45 pm
IlikeJackB wrote:
Mon Jul 09, 2018 10:51 am
The Great Depression: A Diary by Benjamin Roth
[...] Just to mention one point, it's the first time I ever realized a) that even banks that did not fail often froze withdrawals, and that b) as a result there was a market in "bankbooks," with prices published in newspapers. If you had a bankbook showing a $1,000 balance in a bank that had frozen withdrawals, people would pay you varying amounts (like $0.50 or $0.25 on the dollar) for you to sign your bankbook over to them. They would then wait for the bank either to fold, or to start honoring withdrawals again.

User avatar
Gabriel A. Lozada
Posts: 37
Joined: Wed Oct 16, 2019 10:53 pm
Location: Dept. of Economics, University of Utah, Salt Lake City, Utah, USA 84112
Contact:

Re: Composition and safety of TIAA traditional annuity

Post by Gabriel A. Lozada » Sun Feb 16, 2020 10:29 am

Valuethinker wrote:
Fri Feb 14, 2020 10:42 am
Gabriel A. Lozada wrote:
Thu Oct 17, 2019 2:36 pm
Trurl Klapaucius wrote:
Thu Oct 17, 2019 12:16 pm
[...] I come up with a weighted average credit quality of 0.969, which corresponds to a loss rate of 1 – 0.969 = 0.031, which is between that of BBB (1.19%) and BB (6.90%).
It's hard to know what to do with the NR ("not rated") bonds. One perfectly reasonable strategy is the one Trurl Klapaucius takes, which is to ignore them. However, technically, then the 0.9690 sum of the "weighted credit quality" only represents the quality of 100% - 1.15% = 98.85% of the holdings, so the average credit quality of that 98.85% of the holdings is 0.9690/.9885 = 0.9803 (actually, 0.9802 with no intermediate roundings). That's still between BBB and BB, closer to BBB than 0.9690 was, but not as close as the year-end 2015 portfolio was. Only if one were to assume a 100% loss rate on NR bonds would the average credit quality be as low as 0.9690.

Such technical details aside, we've learned from Trurl Klapaucius is that TIAA's General Account quality is somewhat worse now than it was at year-end 2015.
Not Rated bonds are perhaps private placement bonds?

By accepting lower liquidity the investor can improve upon the yield (the costs for the issuer are lower).

I would conjecture that these are investment grade, and a rating of BBB or BBB+ would be the most appropriate (since BBB is the rating of 48% or so of all non-financial corporate bonds, probably use that rating to be conservative; or, alternatively, just assume that the ratings distribution mirrors the rest of the portfolio?).
The question of what to do about Not Rated bonds arises when trying to compute TIAA's average quality from sources such as the one cited by student in his/her post of Oct. 17, 2019, 9:32am, namely https://www.tiaa.org/public/pdf/perform ... rength.pdf. That question however does not arise using the source given on p. 6 of my paper, namely page SI07 (PDF p. 156) rows 9.1 to 9.6 column 7 of https://www.tiaa.org/public/pdf/tiaa_an ... t_2015.pdf, which assigns a rating to every one of TIAA's bonds, both "publicly traded" (column 10) and "privately placed" (column 11). Perhaps TIAA itself, or its accountants, assign (NAIC) ratings to holdings which were "not rated" by any of the credit rating agencies.

The substantive question raised by this discussion is to what extent the quality of TIAA's General Account has changed since year-end 2015. The General Account's year-end 2018 annual report is at https://www.tiaa.org/public/pdf/TIAA_an ... t_2018.pdf and if one repeats the calculations of pages 6 and 7 of my paper referenced above (http://content.csbs.utah.edu/~lozada/Re ... IAA_BH.pdf) for the 2018 data one obtains a quality (one minus the loss rate) of 0.9822, corresponding to an increase in the loss rate from 0.0126 to 0.0178. On my paper's page 8's Figure 1 the horizontal coordinate of the diamond representing the General Account would move slightly to the left, to between the current positions of the letters "n" and "t" of the word "Account." To put this into perspective, on page 12, the entries in row 4 ("HY Corp") of Table 3's matching portfolios would all rise from 6% or 7% to 11%, and on page 13, the entries in row 4 ("CorpHY") of Table 4's matching portfolios would all rise from 6% or 7% to 10%.

student
Posts: 4515
Joined: Fri Apr 03, 2015 6:58 am

Re: Composition and safety of TIAA traditional annuity

Post by student » Sun Feb 16, 2020 10:34 am

George Kaplan wrote:
Sun Feb 16, 2020 1:42 am
Agreed. To my mind, the biggest risk to TIAA Trad is the contractual right that CREF people have to transfer into Trad. Say int rates fell to -2% (not out of the realm of possibility). CREF money in 3% min contracts (RA, SRA) could walk right in. I'm not sure TIAA has any protection against this.
I assume that they can stop letting people contribute to the contract. I have not read the contract so I do not know but I think having such a clause is reasonable.

columbia
Posts: 2288
Joined: Tue Aug 27, 2013 5:30 am

Re: Composition and safety of TIAA traditional annuity

Post by columbia » Sun Feb 16, 2020 10:40 am

Oregano wrote:
Fri Feb 14, 2020 5:19 pm
The special sauce of TIAA Traditional is that it is over-collateralized (i.e., assets exceed liabilities.) That's it. That's why it is rated AAA and that's why it can pay above market rates.

Thus, I don't think credit risk is a large concern. I also don't think rising interest rates are a terribly large concern, either. The long-term viability of the product will be at risk if interest rates stay low or go lower (and my hunch is that is the case). This will become a problem because some of their products have 3% guarantees and the cash flows from even the overcollateralized portfolio won't support that (they might not already). Many vintages are paying much more than 3%. Due to the lock ups to withdraw money and the fact that they are always getting new inflows, TIAA has a lot of flexibility to manage this over time. They would address permanently low rates by cutting interest rates on existing vintages and freezing contributions to products that have 3% guarantees. I would guess they will be able to pull this off so that owners principal value is not at risk over time, but the substantial excess yields that some vintages are currently earning are probably not going to last much longer. Enjoy them while you can.
Most of my Trad. RC is churning along at 4.25%, because of the timing of when I did my initial lump sum into the arrangement.
If you leave your head in the sand for too long, you might get run over by a Jeep.

User avatar
Gabriel A. Lozada
Posts: 37
Joined: Wed Oct 16, 2019 10:53 pm
Location: Dept. of Economics, University of Utah, Salt Lake City, Utah, USA 84112
Contact:

Re: Composition and safety of TIAA traditional annuity

Post by Gabriel A. Lozada » Sun Feb 16, 2020 11:21 am

columbia wrote:
Sun Feb 16, 2020 10:40 am
Most of my Trad. RC is churning along at 4.25%, because of the timing of when I did my initial lump sum into the arrangement.
One can find the current crediting rates on all vintages of the various TIAA Traditional contracts by going to https://www.tiaa.org/public/investment-performance and viewing the "Fixed Annuities" section.

columbia
Posts: 2288
Joined: Tue Aug 27, 2013 5:30 am

Re: Composition and safety of TIAA traditional annuity

Post by columbia » Sun Feb 16, 2020 11:26 am

Gabriel A. Lozada wrote:
Sun Feb 16, 2020 11:21 am
columbia wrote:
Sun Feb 16, 2020 10:40 am
Most of my Trad. RC is churning along at 4.25%, because of the timing of when I did my initial lump sum into the arrangement.
One can find the current crediting rates on all vintages of the various TIAA Traditional contracts by going to https://www.tiaa.org/public/investment-performance and viewing the "Fixed Annuities" section.
Yeah, RC has creeped back to 3.75% for the current contributions. That it’s gone up 50 basis points since September is interesting.
If you leave your head in the sand for too long, you might get run over by a Jeep.

student
Posts: 4515
Joined: Fri Apr 03, 2015 6:58 am

Re: Composition and safety of TIAA traditional annuity

Post by student » Sun Feb 16, 2020 11:30 am

columbia wrote:
Sun Feb 16, 2020 11:26 am
Gabriel A. Lozada wrote:
Sun Feb 16, 2020 11:21 am
columbia wrote:
Sun Feb 16, 2020 10:40 am
Most of my Trad. RC is churning along at 4.25%, because of the timing of when I did my initial lump sum into the arrangement.
One can find the current crediting rates on all vintages of the various TIAA Traditional contracts by going to https://www.tiaa.org/public/investment-performance and viewing the "Fixed Annuities" section.
Yeah, RC has creeped back to 3.75% for the current contributions. That it’s gone up 50 basis points since September is interesting.
It is interesting that RC, the supposedly "bad cousin" with the lower guarantee rate, is having the highest rate (and I think since inception).

User avatar
Gabriel A. Lozada
Posts: 37
Joined: Wed Oct 16, 2019 10:53 pm
Location: Dept. of Economics, University of Utah, Salt Lake City, Utah, USA 84112
Contact:

Re: Composition and safety of TIAA traditional annuity

Post by Gabriel A. Lozada » Sun Feb 16, 2020 2:28 pm

student wrote:
Sun Feb 16, 2020 11:30 am
columbia wrote:
Sun Feb 16, 2020 11:26 am
Gabriel A. Lozada wrote:
Sun Feb 16, 2020 11:21 am
columbia wrote:
Sun Feb 16, 2020 10:40 am
Most of my Trad. RC is churning along at 4.25%, because of the timing of when I did my initial lump sum into the arrangement.
One can find the current crediting rates on all vintages of the various TIAA Traditional contracts by going to https://www.tiaa.org/public/investment-performance and viewing the "Fixed Annuities" section.
Yeah, RC has creeped back to 3.75% for the current contributions. That it’s gone up 50 basis points since September is interesting.
It is interesting that RC, the supposedly "bad cousin" with the lower guarantee rate, is having the highest rate (and I think since inception).
Yes, that is interesting. Its average annual ten-year rate of return is 4.24%, while the RA/GRA figure is 3.99%.

For comparison, the "2m" matching portfolio of my paper (44% Vanguard Long-term Investment Grade Fund and 56% Vanguard GNMA Fund) currently shows an average annual ten-year return of 5.3%, and the paper's "4mt" matching portfolio (34% Long Investment Grade, 28% GNMA, 32% Vanguard Total Bond Market Index Fund, and 6% Vanguard High Yield Corporate Bond Fund) currently shows an average annual ten-year return of 5.2%. Of course, such current ("point-in-time") comparisons do not show much; my paper investigates similar comparisons serially over several decades.

Oregano
Posts: 19
Joined: Fri Nov 22, 2019 9:30 pm

Re: Composition and safety of TIAA traditional annuity

Post by Oregano » Mon Feb 17, 2020 3:53 pm

student wrote:
Sun Feb 16, 2020 11:30 am
columbia wrote:
Sun Feb 16, 2020 11:26 am
Gabriel A. Lozada wrote:
Sun Feb 16, 2020 11:21 am
columbia wrote:
Sun Feb 16, 2020 10:40 am
Most of my Trad. RC is churning along at 4.25%, because of the timing of when I did my initial lump sum into the arrangement.
One can find the current crediting rates on all vintages of the various TIAA Traditional contracts by going to https://www.tiaa.org/public/investment-performance and viewing the "Fixed Annuities" section.
Yeah, RC has creeped back to 3.75% for the current contributions. That it’s gone up 50 basis points since September is interesting.
It is interesting that RC, the supposedly "bad cousin" with the lower guarantee rate, is having the highest rate (and I think since inception).
There is greater risk in RC due to the lower minimum rate, so it's reasonable to offer a higher rate (at times, at least; TIAA is continually reviewing flows to try to optimize across all the option). If they have to cut rates deeply at some point, RC will get hit harder since most of the other options can't go below 3%.

Post Reply