what makes an index investable?? please help
you could buy each security in the index separatly. that makes an index investable.
whats an example of an uninvestable index
I’ve seen an example or two in the CFAI texts (either in the text or a question)…off the top of my head i can’t think of a scenario where the index is univestable.
example of a non-investable fund NCREIF represents direct investments in real estate the index is based on appraisal values of a sample of commercial properties owned by large US institutions, there is no way to buy the underlying according to the index weight (you cannot buy 1/10000 of a shopping mall), and thus the index non-investable
I may be wrong, but I’m pretty sure that the liquidity constraints mean that hedge fund indices like Tremont are uninvestible (along with a having host of other bias problems, like backfill bias, so your index values change retroactively when new funds are added to the index, that would make making it investible a little difficult). I wonder if one could create a futures contract based on a hedge fund index and then create a synthetically investible index? Obviously you’d have to make some adjustment for backfill, but maybe it could be done?
From time-to-time people trade derivatives on hedge fund performance. Obviously, hedge fund owners hate it because if someone is going to get the upside from their performance, they want fees. That dynamic stops the derivatives in a variety of ways. There are tons of non-investable indexes out there. In fact, I think nearly all indexes are to some extent non-investable. Offhand, we have a) Economic indexes. You might be able to invest in CPI synthetically. You can’t invest in any way that I know in the Chicago Purchasing Manager’s Index (although all things are possible if the price is right). b) Capital Constrained Indexes. You sort-of could invest in all the hedge funds in a Tremont index. It would cost you hundreds of millions of dollars and you would be investing in a way that deliberately avoided diversification. c) Liquidity problems - Bond indexes with lots of bonds. You just can’t find and buy all the bonds. d) Reinvestment problems - Almost all “total return” indexes have reinvestment assumptions you can’t do, like immediately reinvesting coupon interest or dividends proportionally in the index. e) Trading cost problems - All indexes that I know of assume 0 trading costs for all rebalancing and reinvestment. f) Disappearing security problems - We were discussing the DJ Commodity indexes a few days ago. When a contract expires they just move to the next one out and include the new price in the index instead of the expired price. Try calling your broker and saying “February just went off the boards at 102.5, so I need an August at 102.5 to keep my index investment tracking correctly” and the usual billion other reasons…
When I read this section, I think most index are investable…however, it is the degree of liquidity and trading cost. Yes, if the manager want to invest, he/she can. However, the trading cost maybe to high to compensate the benefit.
high transaction and liquidity costs don’t make an index univestable. you can still invest in it despite high transaction costs. If you want to depends on other analysis (which means that you likely won’t as costs are too high). But you can still buy the underslying assets if you want to pay.
Not really… I would say that an investable index means you can get index returns. If you are always paying transactions costs so you can’t get those returns, in my book that makes it uninvestable.
the definition of an investable index is not that you can obtain index returns, it is that you can purchase all the securities that make up the index separately and if you want that you can passively hold the index. transaction costs are not part of the definition.
If economic indicator is counted as indices then I have to agree with JDV. My definition for investable indices is simply to have exposure to an index via any means. I would definitely take transaction costs into consideration, at the end of the day, you look at your totally return for an investment hence investable indices. In short, you can invest in NCREIF. Merrill Lynch has a property derivative that has been trade in the market for a while, they are starting to establish their ground in Australia, saw them few weeks ago. It’s a total return swap for NCREIF yield and LIBOR. Obviously, you can use some other rates… Also hedge fund index like Tremont is also investable. Credit Suisse has CS/Tremont index tracker. It is a strategy available to both retail and insto clients with daily liquidity - particularly attractive for people wants diversify exposure in hedge fund but doesn’t like the illiquidity.
So what is the deal with the CS/Tremont index? I tried to figure out how to invest in it and got lots of “No Information Available” (possibly because I answered all questions honestly which says I have < 10M to invest with them). Here’s the thing - Hedge fund NAV’s are not the same as equity prices or mutual fund NAV’s. They are something like a hybrid of company earnings and mutual fund NAV’s with phenomena like “big bath”, gamma, illiquidity recognition or not, etc… In particular, hedge fund NAV’s demonstrate lots of plain-vanilla autocorrelation depending on the fund type. If you can actively trade hedge fund indices without huge transaction costs, your LII stuff should show you how to make a mint doing it. Which, of course, is why it’s hard to do.
Striker, that was my understanding too. I thought “investable” means can you actually phycially buy the asset held in an index; regardless of the trading cost.
I think investable means that there is a passive (i.e. you can automate it) mechanism that produces the same return results as the index. If you hold to this definition strictly, there is no truly investable index because of transaction costs, but most people will give you a little slack for transaction costs that are minor. However, I am not quoting anything that is CFA kosher, so if I’m wrong, please correct.
in the CFA definition, once again, transaction costs have nothing to do when making the decision if an index is investable or not.
Hey JDV, From my understanding, in US, it’s an insto investment vehicle and somehow they are able to provide retail offering in Australia. You are correct that transaction cost is higher but you can always waive the buy spread under some terms and conditions e.g. investment amount of greater than $5m. I understand about the phenomna you’re talking about and I must admit that I did not look into it. I am about to construct my first model portfolio and it just happen that I met them and like the exposure and daily liquidity (I’m still reading on how they do it). To me, this is enough for me to allocate like 5~10% into them. But of course, will need to do my study before recommend them. If you are interested, drop me an e-mail. I promised the PM that I’ll get back to him with lots questions. With your help, I may be able to make myself look more sophisticaed than I really am… slow.takumi AT gmail DOT com
several valid points has been made here and I have learned a lot from you guys. I still have this question if there is an essay question on the level III exam “comment on whether the NCREIF index is investable”, and I answered with “yes, because there are currently derivatives available to invest in the index”. Will I get points for the answer? The CFAI textbook noted that the NCREIT index is uninvestable. But I agree with JDV that all things are possible if the prices are right. Should we agree with takumi’s definition that “investable” means exposure through any means? Will CFA grader agree with that definition?
I don’t exacty know the answer to your question so let me brush it aside. On the CFA exam they are not going to ask you a question that relies on defining some fuzzy term like this. The question they would ask is something like: “Jane Smith is investigating the returns of REIT managers and states to her superiors that all REIT managers suck because they can’t beat the NCREIF index. Comment on Jane’s opinion of REIT managers including her choice of benchmark”. I think ‘investable’ does not depend on the availability of derivatives, especially things like total return swaps. If it did, then everything is investable if you can find someone else to bet on the other side. There is a general theorem about this in finance - “If there is an index available, then Joey will do a total return swap on that index for LIBOR + x for some x” Thus, if investable relies on the availability of derivatives, all indexes are investable by the well-known “Joey Swap Theorem”. Your answer to the above question could include something like a) The NCREIF index is an inappropriate benchmark because it is not ‘investable’ since an investment manager cannot earn the index returns by investing in the underlying components. In this case, the underlying components are not for sale at the index prices, the index does not include the considerable transactions costs in purchasing the components, and the cost of purchasing all the components of the index is way beyond the resurces of any of the PM’s.
JDV thanks very much for taking the time to provide to the point response. Regarding the question I previously phrased, my feeling is also that CFA won’t ask such a fuzzy question. Your asnwer is reassuring. Indeed, “Joey swap” would make all indicies investable. If such derivative is available, the scale would be very limited that PM cannot regard it as a generally applicable instrument. Your hypotehtical question really helps get the feeling on what kind of question may be aksed in the CFA exam. I think your answer is definitive in this “investable index” question.