— Paul Amery's Blog

In a famous cartoon from a 1993 issue of New Yorker magazine, two dogs are chatting, one sat on a chair behind a computer terminal.

“On the internet, nobody knows you’re a dog,” he tells his canine companion.

The idea, of course, is that all internet users have an equal voice and equal access to information.

But in her excellent “Weapons of Math Destruction”, data scientist and former hedge fund manager Cathy O’Neill argues that the reverse is now the case. We are far from equal. We are ranked, categorised and scored in hundreds of models by powerful algorithms in ways we may struggle to imagine, let alone understand.

Often the output of computer programmes might seem little more than an inconvenience. If we perform a search for a new pair of shoes or a holiday, ads start to appear on the websites we visit, promoting similar products. There’s the urban myth that airline websites raise prices when you make a repeat visit—which, if true, is more annoying.

But what O’Neill describes in her book is a far more pervasive and destructive use of algorithms.

The growth of for-profit colleges in the US, which she describes as “diploma mills underwritten by government-financed loans”, has been driven by predatory targeted advertising, she shows, aided and abetted by the internet giants.

Corinthian Colleges, a company which filed for bankruptcy in the US last year, spent $120 million on marketing annually, much of it to generate and pursue 2.4 million potential leads. This led to sixty thousand new students and $600 million a year in annual revenue, O’Neill notes.

Via ads driven by secret algorithms, companies like Google and Facebook helped the colleges target people in particular geographical locations highlighted as low-income. Those who had taken out high-interest loans or fought abroad (in the US, war veterans can easily obtain government funding for tuition) moved even higher up the list of potential victims.

Sometimes “lead generators” working on commission would even post fictitious job ads on the net, purely for the purpose of harvesting personal details. The details could then be sold to the colleges’ recruitment teams, who would pay $85 for each potential target. Some of those unwittingly snared in this way would then receive up to 120 cold calls a month.

Disgraceful? Yes. Illegal—not yet, at least in the US. But O’Neill makes a convincing case that the misuse of algorithms is reinforcing social inequalities and, worse, subverting democracy on a worldwide basis.

Far from fighting political campaigns based on a single published manifesto, parties or candidates seeking election now place targeted messages with particular groups of voters, often hiding them from the broader public.

Ted Cruz, candidate for the Republican party’s presidential nomination, did this in Florida last year, writes O’Neill, showing web-based ads to a meeting of the Republican Jewish Coalition at a hotel in Las Vegas. The ads, which reiterated Cruz’s commitment to Israel and its defence, were visible only to those located inside the hotel. Weren’t the rest of the US public entitled to know what he was promising?

O’Neill paints an alarming picture of how social media news feeds could be manipulated to game the political system by playing with voters’ emotions on election days. Facebook, for example, has enormous power to affect what we learn, how we feel and whether we vote, she writes.

The vicious cycles created by algorithms extend to the US judicial system (in certain states, computer-generated scores drive longer sentences for those classified as at a higher risk of reoffending), the job market (anyone with a past mental health issue may be blackballed) and insurance (microtargeting means a move away from the principle of risk sharing, to the benefit of the company but not the consumer, who may face massive rises in premiums).

All this adds up to an inversion of the American national motto of “E Pluribus Unum”, (“Out of Many, One”), argues O’Neill.

“Weapons of math destruction reverse the equation,” she writes. “Working in darkness, they carve one into many, while hiding from us the harms they inflict upon our neighbours near and far.”

What’s the solution? Greater openness for data models, a moral code for programmers, a shift towards the European model of opt-ins for the reuse of personal data: all these may have some impact, the author suggests.

A more robust policy—making sure that social media companies open up their algorithms to third-party audits to detect potential biases—has so far been resisted by the web giants. Given the firms’ power, regulators have their hands full.

The internet, computers, social media and big data are not going away: they are an ever more integral part of our lives. But Cathy O’Neill makes a convincing case that algorithms have run amok in secrecy, and that the code with which they operate needs to be tamed.

Read More

For the 1929 Wall Street crash to turn into a global depression, one extra ingredient was needed: protectionism.

The Smoot-Hawley tariff, applied in June 1930 by the US government to agricultural imports, set in train a cycle of tit-for-tat measures that brought global trade down by a half in two years.

The succession of events that followed is well-known: rising poverty and unemployment, growing left and right-wing extremism throughout Europe, international tensions and, eventually, war.

The pro-Brexit result of last week’s UK referendum is the closest 21st century equivalent to Smoot-Hawley we have yet seen. The symbolic importance of the imminent withdrawal of the world’s fifth-largest economy from its largest tariff-free trading zone is hard to exaggerate.

This shock event has taken place against a rapidly deteriorating outlook for world trade and may well signal the acceleration of the trend. According to a sobering recent research note from Insight Investment, protectionist and discriminatory measures are becoming more commonplace.

The US has introduced 43 trade barriers to protect its steel industry in the past seven years, says Insight, recently culminating in a new tariff on Chinese steel. Over the same time period Europe has introduced 14 new anti-dumping duties.

Last week, the World Trade Organisation (WTO) reported that, between October 2015 and May 2016, members of the G20 group of leading industrialised nations implemented new protectionist measures at the fastest rate since the WTO began keeping records in 2009.

During this period, the WTO said, the G20 economies applied 145 new trade-restrictive measures, equating to an average of almost 21 new measures per month.

Insight’s CEO, Abdallah Nauphal, cites alarming demographic trends and the growth in debt levels as other factors threatening to rend politics more insular and international cooperation much more difficult over coming years.

Global trade volumes, which for decades have risen at a faster rate than GDP, have lagged economic growth rates ever since the financial crisis and are now in decline: the value of global trade fell 14 percent in dollar terms last year. The Baltic Dry index, a barometer of the global shipping market, is bumping along at multi-year lows.

According to a new article by Jeffrey Rothfeder in the New Yorker, the average charter rate for the largest cargo carriers has fallen from two hundred and fifty thousand dollars in 2008 to less than three thousand US dollars per day now, reflecting massive overcapacity in shipping and falling demand. That’s a stunning, 99 percent decline in eight years.

Signalling rising protectionist sentiment, one of the ideological leaders of the Brexit campaign, Chris Grayling, threatened this weekend to restrict agricultural imports from France to the UK if Paris or Frankfurt tries to “emasculate” the UK’s financial sector.

The right-wing members of the Brexit campaign have long agitated for immigration controls, step two in an attack on the EU’s basic principles of free movement for goods, labour and capital.

Boris Johnson, the putative leader of the next Conservative government, sought to play down such fears this weekend, but only by making the absurd claim that UK citizens would be able to enjoy access to the rest of the EU, whether to work or live, while the UK would be able to control movement in the other direction via an immigration points system.

If restrictions on the free movement of goods and people spread across the region and worldwide, limits on capital flows will surely follow. Taken together, these trends threaten an economic recession on a scale not seen since the 1930s.

Brexit is likely to go down in economic and political history as signalling the start of a very dark period.

Read More

US mutual fund manager Third Avenue Funds’ justified its decision on Wednesday to gate redemptions from its Focused Credit Fund (FCF) as follows:

“We believe that, with time, FCF would have been able to realize investment returns in the normal course. Investor requests for redemption, however, in addition to the general reduction of liquidity in the fixed income markets, have made it impracticable for FCF going forward to create sufficient cash to pay anticipated redemptions without resorting to sales at prices that would unfairly disadvantage the remaining shareholders.”

Instead, investors remaining in the fund will be placed in a Liquidating Trust and will have to wait for the return of their money until the fund manager can dispose of the assets at what it calls “reasonable prices”. That may take months or more.

US mutual fund investors entering and exiting a fund do so on the basis of the fund’s net asset value (NAV), which is hard to determine in illiquid markets. If exiting investors are able to do so at prices that do not reflect the liquidity of the underlying market, they obtain an unfair advantage over those remaining in the fund, whose interests are diluted.

A central question is unanswered in Third Avenue Funds’ statement. Why not mark down the fund’s NAV until it reflects the price at which its portfolio of junk bonds can be sold?

Marking fund NAVs to the “bid” side of quotes in the portfolio is common practice in a bond fund, since it represents the most conservative approach. A portfolio of assets is realistically worth what it can be sold at.

The fact that Third Avenue felt unable to do this is the most worrying aspect of its fund closure. There are persistent reports that liquidity in junk bonds is so poor that a request to market makers for a quote on a portfolio can be left unanswered. Bonds go “no-bid”, in other words. As a former market maker in emerging market debt, I remember from the Russian crisis of 1998 that the market could simply stop and there was no prospect of selling your holdings at all.

The risk of bond illiquidity reoccurring on its scale has been masked by central banks’ zero interest rate policies. As many have argued, putting illiquid, higher-yielding assets in funds promising daily liquidity to investors has created large-scale structural risk. In simple terms, it’s been an accident waiting to happen. Following Third Avenue Funds’ decision to suspend redemptions, the managers of other funds investing in junk bonds, bank loans and emerging market debt will be looking nervously over their shoulders.

Read More

The widely used distinction between passively and actively managed funds dates to the introduction of index funds in the 1970s. Similarly, finance courses teach students that investment portfolios can be separated into beta and alpha components.

Beta is the return you get through systematic exposure to an asset class as a whole. It’s accessible by purchasing an index fund or ETF based on the capitalisation-weighted index. Alpha is then the excess return targeted by active managers. It’s long been standard practice among professional investors to use this mental shortcut and allocate assets accordingly when designing portfolios.

But these simple categorisations don’t help us when it comes to smart beta, the label commonly attached to rules-based investment strategies.

If you want an equity strategy focused on US value stocks, high-momentum shares or quality Japanese companies, there’s a range of smart beta indices available for purchase off the shelf. Many ETF issuers include index-tracking smart beta funds of this type in their product ranges.

For example, iShares’ European ETF range includes funds focused on size, momentum, quality and value factors, all tracking indices from MSCI. Lyxor offers smart beta ETFs targeting quality income, minimum variance, value beta and a number of market factors, based on indices from Société Générale, FTSE Russell and JP Morgan.

So why, when launching a range of similar strategies as ETFs this morning, has Vanguard decided to sidestep the index part of the strategy altogether?

Vanguard’s new ETFs—global liquidity factor, global minimum volatility, global momentum factor and global value factor—are instead labelled as active.

Although these funds offer exposure to essentially the same strategies as index-based ETFs from other providers, Vanguard says its new ETFs will not track an index and there’s no mention of smart beta in the firm’s marketing literature.

Instead, the new Vanguard ETFs use the (cap-weighted) market index as a reference benchmark, while the fund manager’s investment objective is long-term capital appreciation, rather than tracking or relative outperformance.

Vanguard’s approach is consistent with a research paper it published this summer, in which it suggested that smart beta indices should be labelled as rules-based active strategies. The firm has long argued that index-based investing should involve cap-weighted benchmarks and nothing else.

Are these semantic differences important? Perhaps not. But I take Vanguard’s move as a sign of an intensifying turf war between asset managers and index firms over smart beta.

Index firms argue that their smart beta strategies operate on the basis of transparent index rules, providing reassurance to investors that a tracker fund is not going to diverge hugely in performance from what the index does. There are plenty of examples of quantitative investment managers getting their models wrong and blowing up.

Asset managers can claim that they have the quantitative investment expertise to produce smart beta in-house, and that by managing the money themselves they are freed from the obligation to rebalance their portfolios according to a set schedule, something that exposes index-based funds to front-running. And avoiding the use of an index in a smart beta strategy means saving on the licensing fees paid to the index provider.

In Vanguard’s case, the firm is clearly betting that its own name carries more weight amongst potential fund purchasers than the name of an index firm attached to the product.

Regulators may yet become more involved in the competitive battle between fund and index firms. There’s long been a legal loophole by means of which asset managers are prohibited from advertising products that use back-tested performance data, whereas index firms—since they are not treated as product promoters—may do so when describing how indices work.

If you think smart beta is a good thing, rather than a marketing gimmick, this turf war is good news, since fees are coming down. Vanguard’s new factor ETFs cost 0.22 percent a year, undercutting similar strategy ETFs from other providers. But the increasing confusion over terminology means navigating your way around competing investment products isn’t getting any easier.

Read More

When you agree a trade with a counterparty you’re done, filled, executed, correct? My word is my bond, and all that?

Depending on where you do your trading business, yes, possibly, or not at all. In the worst case, you may have granted your counterparty a free option to reprice trades to its advantage.

Following last week’s dramatic breakdown of the Swiss franc/euro currency peg, several electronic FX brokers decided to “revisit” executed client orders.

For example, according to commenters on the Bronte Capital blog and elsewhere on the internet, Saxo bank told its clients that trades agreed and confirmed in the immediate aftermath of the currency move would be moved to a level more favourable to the broker.

One client’s apparently executed stop-loss at EurChf 1.20 was moved to 0.9625, another’s from 1.184 to 0.9625. For leveraged clients, such a repricing may have turned a painful shock into insolvency.

How are Saxo-like repricings possible?

It turns out that the Danish bank had covered itself via the legal small print of its client contracts, which stated:

“It is possible that errors may occur in the prices of transactions quoted by Saxo Bank. In such circumstances, without prejudice to any rights it may have under Danish law, Saxo Bank shall not be bound by any Contract which purports to have been made (whether or not confirmed by Saxo Bank) at a price which: 

(i) Saxo Bank is able to substantiate to the Client was manifestly incorrect at the time of the transaction; or 

(ii) was, or ought to have reasonably been known by the Client to be incorrect at the time of the transaction. 

In which case Saxo Bank reserves the right to either 1) cancel the trade all together or 2) correct the erroneous price at which the trade was done to either the price at which Saxo Bank hedged the trade or alternatively to the historic correct market price.”

But surely exchange-based trades—where the exchange stands between buyer and seller—aren’t vulnerable to such after-the-fact repricings or cancellations?

Unfortunately they are.

In 2013 Goldman Sachs was able to bust (get out of) a large number of options trades which it had priced incorrectly, apparently as the result of a computer error.

But why was Goldman able to do so, when in a similar case a year earlier market maker Knight incurred huge losses and was forced into a fire sale of its business?

Because the rules of NYSE Euronext’s options trading venue, where most of Goldman’s losing options deals were apparently executed, permitted it.

By contrast, cancellation trades for equities (imposed after the 2010 “Flash Crash”, another case of trades being torn up after the event on the basis of an arbitrary judgement by the exchanges involved), were apparently stricter, meaning that most of the trades spewed out by Knight’s rogue algorithm were forced to stand.

But it gets even more complicated. According to Bloomberg, while NYSE Euronext voided its share of Goldman’s options trades, the bank’s options trades executed on other exchanges were either left untouched or repriced:

“The majority of the reviewed trades at the International Securities Exchange were adjusted and not canceled, based on the exchange’s rules, Bats Global Markets Inc. let the few hundred trades on that venue stand while Boston Options Exchanges adjusted trades.”

“It’s crazy when you look at how ISE does it one way, CBOE does it another way, Amex does it yet another way, that is absurd and it is demonstrably dangerous,” Mark Longo, founder of the Options Insider, told Bloomberg at the time.

“Let’s say you’re a market maker and you put up the same trade on three different exchanges, then you have three different processes you have to deal with to bust or adjust the exact same trade and you may have three different outcomes. How is that beneficial to anybody?”

Good question. And in the aftermath of another algorithm-accelerated rout in the Swiss Franc last week, it’s one that every investor should be asking. Understand how the rules of your trading venues or counterparties work in a market crisis or you risk being the next victim.

Read More

Today the main US-listed high yield bond ETFs, BlackRock/iShares’ HYG and State Street’s JNK, are trading at levels unseen for over two years. At the time of writing, HYG is at 88.10, down nearly seven points in price from an interim peak of 95 in late October.

Junk bond indices have been hit by the sharp post-summer decline in oil prices and fears of broader deflation: according to Dave Nadig of ETF.com, 15% of the index underlying HYG is in bonds issued by energy companies; for JNK’s index it’s 17%.

Here’s a three-year chart of HYG, JNK and Invesco Powershares’ BKLN, another high yield ETF, this time investing in senior bank loans from sub-investment grade companies. The three ETFs’ prices are rebased to 100 at the end of 2011.

hyg jnk bkln

Source: Google Finance

Over recent days many analysts have been pointing out the divergence between the weakness in the junk bond market and the relative strength of US equities. In the past such divergences have hinted at equity market falls to come, since the credit market tends to act as a leading indicator.

Unlike during previous bouts of market nervousness, the price declines in the high yield market haven’t so far been met with large-scale ETF redemptions. During the first half of 2013, the last significant sell-off, HYG and JNK lost up to 20% of their net assets, a collective outflow of $5 billion. This time round things have so far been different: Nadig noted earlier this week that the two funds have pulled in almost $3 billion in new assets so far in the fourth quarter. Of the three ETFs in the chart above, only BKLN has seen net redemptions for the quarter to date.

Here are the Q4 fund flows for the three ETFs, courtesy of ETF.com.

HYG flows Q4 JNK flows Q4BKLN flows Q4

That’s what’s concerning about recent price weakness in the sector. During previous market downturns (notably in June last year) the redemption mechanism of high yield ETFs has been put under strain. Meanwhile, by many accounts, the underlying liquidity of the corporate bond markets has worsened further.

What if the sell-off in junk bonds hasn’t even really started?

Read More

Take two exchange-traded funds (ETFs), both on sale to retail investors, both investing in dollar-denominated high yield bonds. The funds are run by the same asset management firm, both track an index from the same benchmark provider and both promise their authorised participants (APs) the ability to create and redeem fund units daily on the basis of the underlying net asset value (NAV).

APs are the entities that transact in wholesale lots with an ETF fund issuer, an interaction that dictates how the fund is priced in the secondary market, where most of us buy and sell it.

Yet in one prospectus we read that the ability of investors to exit the fund may be curtailed in a number of circumstances. The ability of APs to redeem their fund units, says the prospectus, may be suspended if any of the principal underlying markets are closed or if the fund’s directors decide it’s difficult to determine the NAV.

More broadly, redemptions can be halted if the directors decide a suspension of dealing is in the interests of the fund, its shareholders or the investment company (an “umbrella” structure under which tens of ETFs are issued).

And even if none of these things occur, the fund may limit redemptions to 10% of the fund’s assets a day, meaning that large withdrawals may be scaled down and that those heading for the exits may have to wait.

Under any of these circumstances you could expect secondary market trading in the ETF to come to an effective halt.

In the other prospectus—remember, for a fund sold by the same firm, tracking an index in exactly the same asset class—there’s no mention of a possible suspension of redemptions.

All we read is that “if particular investments are difficult to purchase or sell, this can reduce the fund’s returns because the fund may be unable to transact at advantageous times or prices” and that, if it’s difficult to obtain reliable quotations for securities held by the fund, its manager may use a so-called “fair value” approximation of the securities’ worth in order to calculate the NAV.

Why the difference in language? The first fund operates in Europe and in compliance with the region’s UCITS rules. The second is a US ETF, operating according to the 1940 Investment Company Act, the governing regulation for US mutual funds.

A central principle of the 1940 Act is that investors should be able to redeem their fund units on demand. Unlike in Europe, where local regulators accept the possibility that redemptions may be suspended or “gated”, the US mutual fund rules dictate that fund inflows and outflows should continue even when the liquidity of underlying markets is compromised. There are very narrow exceptions to this principle, mainly relating to the closure of local (US) equity markets.

Does this difference in regional approaches matter?

In its March 2013 principles for the management of liquidity risk in collective investment schemes, IOSCO, the international coordinating body for securities regulators, accepted that liquidity crises in funds are less likely to cause systemic confidence problems than when the same occurs in the banking sector. Investors know they can lose money when buying a fund, or they should do.

But IOSCO’s principles are very high-level and non-prescriptive, no doubt as a result of the transatlantic differences in fund frameworks. In 2012 the regulatory body skirted around the fundamental gap in mutual fund rules by saying that the suspension of redemptions by a mutual fund may be justified only if permitted by (local) law.

IOSCO—whose principles are non-binding—has advised fund managers that they should not promise more frequent liquidity to investors than is appropriate for the underlying asset class and that liquidity risk and a fund’s liquidity risk management process should be effectively disclosed to prospective investors.

Again, this is subject to very wide interpretation in practice. Mutual funds promising daily redemptions (and, in the case of ETFs, instantaneous dealing), now invest on a large scale in a variety of asset classes that have been prone to liquidity crises in the past, from high yield bonds, to emerging markets equity and debt and even senior bank loans.

BlackRock, the asset manager responsible for the two ETFs I mentioned earlier (Europe-listed SHYU and US-listed HYG), has recently called for globally consistent best practices for fund structures, liquidity risk management and investor disclosure.

This is a hot topic, with those charged with ensuring the stability of the financial system now taking a much closer look at whether mutual fund run risk could cause wider contagion. IOSCO and the G-20 Financial Stability Board are due to issue a new consultation on systemically important non-bank financial institutions by the end of the month.

If a fund issuer gates redemptions in one jurisdiction, could there be a run from its funds elsewhere? I don’t know. But while local rules continue to differ so markedly, it’s hard to see how US investors are getting the same message on fund liquidity risk as those in Europe.

Read More

The column inches devoted to castigating bankers for benchmark abuse pile up. But the clients who gave banks the fixing orders in the first place bear some responsibility too.

There’s a common theme running through the benchmark abuse cases: Libor and FX. Banks were given orders to execute client trades at levels that referenced these market benchmarks, while also contributing to the benchmark-setting process.

Clearly, banks should not have been using advance knowledge of clients’ orders to push fixing levels one way or another: this kind of behaviour happened on a large scale, involving collusion between traders at multiple banks.

But what of the clients who gave the order to “trade at the fix” in FX? Were they innocent victims?

If a large corporation or fund manager wants to (say) sell a billion dollars for euro he can trade it in the market in a single order, split it into smaller orders or use an electronic algorithm. All these methods threaten a degree of market impact; in other words, the FX rate moving against the manager as a result of the size of the order.

But (in the past, at least) the fund manager could also hand the order to a bank and instruct it to conduct the whole deal at the day’s 4pm fixing rate.

Although the fixing rate had a bid-offer spread to compensate the market maker for making the trade, the spread was always the same, regardless of the size of the client order and the state of the markets.

There’s a good reason why a fund manager might want to trade at the exact fixing price in a market like FX (or, for example, at the closing index price in an equity market). The fund manager is incentivised (often, formally, via his contract with a client) to minimise tracking error against his own performance benchmark.

Trading at the fix (or at the equity market close) is therefore safer for the fund manager, since it reduces the risk of an unwelcome divergence in performance between fund and benchmark. In some cases clients appear willing to trade at a specific point in time even if this costs money.

It’s unsurprising and no secret that the more money following a particular index or market benchmark, the greater the likelihood of distortions in the index’s or benchmark’s price behaviour.

Fund managers, and other users of the FX fixing, had a right to expect that information regarding their orders would not be abused by the banks.

But the existence of a set spread for “fixing” trades should have raised suspicions that the banks were making money on the side from client orders.

And, unfortunately, by the very act of requiring the banks to commit to execute orders at an (as-yet unknown) fixing rate, irrespective of size, clients acted as the enablers of abusive behaviour.

While the banks are rightfully being brought to task for their role in benchmark abuse, those who “fed” the fix should also recognise they got things wrong.

Read More

US fund manager PIMCO is apparently being investigated by the US regulator for valuing bonds in its Total Return ETF differently to the prices at which it bought them, providing a temporary boost to fund returns as a result.

“The inquiry comes amid escalating scrutiny by the SEC of whether investment funds are valuing assets accurately and fairly,” reports the Wall Street Journal.

In one sense this is hardly a story at all. In another it’s a very big issue. Why?

The reasons why PIMCO’s reported behaviour may not have breached standard market practice are the following.

First, unlike in the equity market, the liquidity of the bond you’re trading depends greatly on the size of the deal.

“On a typical [corporate] bond issue, the bid-offer spread on a US$1 million transaction might be 1 percent  of par value, but for a smaller order the spread might be 3 percent  or more,” a bond trader told me a couple of years ago.

Press reports suggest that PIMCO may have been buying odd lots of bonds (with wide bid-offer spreads) for its ETF and valuing them assuming more standard deal sizes, generating an apparent (though temporary) performance boost to the ETF. This assumes that the fund manager was able to buy its bonds on the “wrong” side of the spread: at or close to the bid price. Unusual, but possible, given PIMCO’s clout.

Second, mutual fund managers’ practices in valuing bonds vary widely and typically allow the managers a great deal of discretion. Yes, managers use external pricing services, but they often retain significant control over pricing policies.

PIMCO’s Total Return ETF prospectus tells us that “portfolio securities…are valued at market value. Market value is generally determined on the basis of last reported sales prices, or if no sales are reported, based on quotes obtained from a quotation reporting system, established market makers, or pricing services,” and that “securities…for which market quotes are not readily available are valued at fair value as determined in good faith by the Board of Trustees. The Board of Trustees has adopted methods for valuing securities and other assets in circumstances where market quotes are not readily available, and has delegated to PIMCO the responsibility for applying the valuation methods.”

So if your bond is relatively illiquid and has a 3% spread between bid and offer prices you have pretty free rein to value it anywhere within that spread, as long as your approach is consistent. This practice may strike outside observers as lax, but it’s how the mutual fund business has long worked.

In 2007, academics from the Mason School of Business published a prize-winning paper showing how the same corporate bonds are priced differently by different mutual fund groups. Some fund firms priced a bond based on secondary market bid prices, others based on the mid-point between bid and offer prices. Around a third of the firms (like PIMCO) refused to tie their pricing policies explicitly to bid or mid prices and instead referred to pricing on the basis of “fair value”, a concept affording substantial discretion to the fund managers.

The academics also produced evidence of return “smoothing”: those funds that had outperformed had a tendency to mark bonds at higher prices, while those that had underperformed tended to do the reverse.

Such smoothing basically cheats one or more of three constituent groups: existing, new, or redeeming fund investors, since someone is dealing at too favourable or too unfavourable a price. But according to the plentiful evidence supplied in the paper, this practice has been going on for years, and it’s not against the rules.

Unless the SEC’s enforcement policies have changed markedly (and I can’t find any evidence for this), I can’t see how it can nab PIMCO for its valuation practices, however loose the “fair value” pricing policy might sound, that have been industry practice for years, as long as the firm was acting consistently and in accordance with the prospectus language.

Cases where the regulator has brought enforcement action against a fund group have tended to involve much more dramatic overstatements of funds’ NAV than those reportedly involved in the PIMCO case.

On a deeper level, this story is of course quite alarming. It reminds us of how deeply illiquid large segments of the bond market really are, even with the recent near-zero levels of interest rates, and how we disguise that illiquidity by putting bonds into daily dealing mutual funds and ETFs and marketing them as more liquid than their constituents. This comment isn’t aimed at PIMCO, by the way.

There’s an irony here, as well. The SEC that’s reportedly investigating PIMCO for overoptimistic valuation policies in a bond fund was long set against reforms forcing money market funds–another type of mutual structure–to value their holdings at market prices, rather than at a notional (and often fictitious) $1 a share. But that’s another story.

Read More

Could the post-LIBOR regulatory crackdown on banks exacerbate price movements when indices are rebalanced?

Late afternoon on Friday shares in Dutch bank ING jumped nearly 5% in heavy trading on the Euronext Amsterdam exchange.

ING Price Spike

ING

Source: Yahoo Finance

The price spike has now been fully reversed. A 5.75% increase in ING’s share price on Friday has been followed by a 4.69% decline yesterday and a further 1.23% fall so far on Tuesday morning.

ING’s late-Friday price move seems related to the reconstitution of equity indices. Friday’s market close set the price at which ING’s shares entered two indices from STOXX—the Euro STOXX banks index and the STOXX Europe 600 Banks. The latter index is a popular one: it underlies several European equity sector ETFs and is reportedly used in many bilateral swap agreements.

Both STOXX and ING declined to comment on Friday’s share price movements.

Price movements in shares entering and exiting benchmarks are a well-known phenomenon, and an unsurprising one. Index changes mean unavoidable cash flows in a whole range of tracker products, including futures, swaps, ETFs and index funds, which other market traders can try to exploit.

STOXX announced the September constituent changes for its indices on August 26 and between August 25 and September 18 ING shares had already risen 8.43%, a not-uncommon price rise between the announcement date for an index addition and the date on which the addition becomes effective.

So why the unusual late-Friday action?

According to one trader I’ve spoken to, post-LIBOR restrictions by banks’ compliance departments on their traders’ index-related deals may be having the effect of forcing market participants into the end-of-day auctions on stock exchanges, which are used to set closing prices.

Previously, said the trader, bank dealers could pre-position their trading books for index changes. Now, he argued, after the LIBOR revelations banks are scared of being seen to exploit benchmark-related trading flows. Instead, he told me, they are now taking the most conservative option: trading at the actual price point at which the index change occurs, even if this leads to a less efficient execution.

Managers of index-tracking funds and financial products face a similar dilemma: place a trade in the closing auction and risk losing money (as anyone buying ING at Friday’s close would have done by Monday morning) or attempt to trade either pre- or post-close. Shunning the closing exchange auction may mean you avoid involvement in a crowded trade, but fund managers will incur tracking error if the fund’s execution price is significantly different from the price used by the benchmark calculator.

Rebalancing-related share price movements may be exacerbated by another factor, the trader told me: the widespread use of volume-weighted average price (VWAP) algorithms. VWAP algos automatically seek to trade when volumes are highest.

“The more volume in the closing auction, the more the VWAP will seek to execute there, leading to more volume in the auction, more VWAP execution….in other words, a feedback loop,” the trader told me.

Not everyone I’ve spoken to agrees with this explanation. The ING price spike could have been caused by someone realising late in the day that they were incorrectly positioned for the index change, or even by a mistake, other traders suggested.

As the volumes of money in tracker products increase, there’s more at stake when indices rebalance. It’s clear that these events involve a trade-off between transparency and making it difficult for those seeking to game index changes. But it would be ironic if, by making trades more crowded, measures taken to clean up the benchmark business had a detrimental effect on the efficiency of indices.

Read More