Thursday, 27 February 2020

Bahrain Middle East Bank FY 2019 Review and Suggested Shareholder Questions for the AGM

Lots of Tunnel Little Evidence of Light

A quick review of BMB’s FY 2019 financials provide no real reason for optimism.  

Based on that review, I’ve included some suggested questions for shareholders attending the upcoming AGM.

FY 2019 Performance Review

As might be expected given the problems BMB is facing, its business activities were adversely affected. For the year net income was a loss of USD 3.6 million.

Other information of note.
  1. Non performing bank obligation: BMB identified a USD 13.1 million non performing obligation from a “locally incorporated” financial institution. I did not see an explicit reference to this in the FY 2018 report. 
  2. Non performing Assets Under Management: As per Note 28 some USD 32.897 million of Assets Under Management as of FYE 2019 (91.5% of total AUM) are placed with related parties. These assets are non-performing. I missed this in the FY 2018 report which shows it pays to read all the notes.
  3. Focus on recovery means little to no focus on income generation: The Board has identified recovery of related party exposures as its primary mission. The Central Bank of Bahrain has also placed restrictions on the bank’s activities. A sensible step given its financial condition.
Questions for the AGM

Shareholders will rightly be focused on the progress and prospects for recovery of related party exposure.

Below are some suggested additional questions.

  1. The long term future of BMB depends on getting new capital. Otherwise the bank will be liquidated. 
  2. Liquidations generally result in less recovery than ongoing concerns. 
  3. Recapitalization is key to a long term future for BMB.
  4. As per BMB’s Articles of Association-which mirror the Bahrain Commercial Companies Law-, the minimum acceptable quorum for an Extraordinary General Meeting requires attendance by shareholders holding 25% of the bank’s shares.
  5. Since ANI owns just under 81% of the bank, it appears that an EGM cannot be held without their active participation. That seems unlikely given the risks such participation would expose them to.
  6. How, if at all, can a recapitalization take place if an EGM can’t be held?

USD 32.897 Million in Non Performing Assets Under Management
  1. Are the assets non-performing because of underlying economics of the investments?
  2. Or was there fraud?
  3. Are the related parties the same entities as with the credit exposure, i.e., the Turkish Three Amigos?
  4. Does the bank have any legal liability to make its customers whole? Note this particular question is likely to be one that the bank may prefer not to answer.
USD 13.1 Million Non Performing Bank Obligation
  1. Is the bank unable to pay BMB?
  2. Or is the bank unwilling to pay BMB perhaps because this is entwined with related party transactions?
  3. The bank appears to be located in Bahrain. Is the Central Bank helping BMB with collection, assuming this is not a case of general insolvency of that bank.
If AA were to attend, you can bet he’d have questions about the single regional financial institution to which BMB owes some USD 127.6 million.

As per the financials (Note 2 and Note 12), it seems clear that that bank is constrained from removing its deposits from BMB.

AA would want to know why this is the case? And of course the identity.  Another question that the bank and its regulator may prefer to avoid.  

Link to earlier post here.

Wednesday, 19 February 2020

Sunday, 16 February 2020

Anti Money Laundering Some Inconvenient "Facts"

Not a Red Arrow in the Quiver

Last month Matthew Collin published a blog entitled “Angola and the money laundering paradox” in which he noted that the dos Santos case highlighted certain paradoxes about money laundering.

His key points were that contrary to what many believe a large volume of money laundering takes place in jurisdictions that score well on Transparency International’s Corruption Perceptions Index, have good ratings (mutual evaluations) on their anti money laundering regulations and system from the FATF or similar bodies, score high on transparency,, etc.

Be sure and read his article it is full of worthwhile insight and information.

None of this is surprising to those involved in international finance nor to those who follow money laundering.  

There are more inconvenient "truths" or at least observations about money laundering. 

As usual, AA is not shy about highlighting them.

First, the "developed and well-regulated" markets are where the bulk of large value money laundering takes place over all three stages of the process: placement, layering, and integration. 

As the chap in the picture above (Robin Hood or Robin Gud) will tell you, the best place for a “hood” to hide out is where there are lots of trees.

It’s much easier to get lost among the trillions of dollars of daily transactions in the major "developed and well-regulated" markets.

These are places that you’d like to hold financial or other assets (properly disguised of course) for a variety of reasons:  greater liquidity, potential for appreciation, systems that protect your rights from arbitrary actions.

Those with assets to sell or banking services to provide in those markets are keenly interested in maximizing sale proceeds or service revenues. And perhaps not of the sort to get overly "fussed" about the source of funds.

A booming property market and vibrant stock market are every government's dream.

This is no secret. There are reliable official estimates of money laundering that cut through imaginary "halos", including those of their own jurisdictions. 

Here are two examples.

Each year the US Department of State publishes the International Narcotics Control Strategy Report pursuant to a statutory requirement. Usually in late March covering the prior year. The INCSR is based on input from various US governmental agencies.

As you'd expect with any national source, it is not free from political considerations. 

Within Volume II of the INSCR is a list of “Major Money Laundering Jurisdictions”. 

You’ll find the list of such countries for 2018 in the INCSR 2019 Volume II pages 14 -15. 

Tucked in among the usual suspects of Afghanistan, Uzbekistan, the UAE, you’ll find the United Kingdom, the United States, etc. 

In past years, the DoS would rank countries as being of “primary concern”, “concern”, and “other countries monitored.”

In the “primary concern” list, the United States, the UK and several European countries were routinely included.  

Here’s an archived version from the 2017 INSCR. 

The UK’s National Crime Agency notes that “Although there are no exact figures there is a realistic possibility that the scale of money laundering impacting the UK annually is in the hundreds of billions of pounds.” 

If you think about a country like Angola, any serious money laundering through the financial system there would be noticeable because of the lower volume of transactions. Not enough trees. In some cases countries only have “bushes". 

Here's an example of how "lack of trees" can aid in identifying unusual, perhaps even suspicious transactions.

Back in 2016 SWIFTs August RMB Tracker noted further imaginary progress in the development of the RMB as an "international currency" citing a spike in RMB transactions in "South Africa".

A fairly cursory examination of publicly available material disclosed that the spike wasn't only in transactions for South African customers.

More importantly that analysis disclosed the spike was primarily due to a single RMB 2.7 billion transaction in Mauritius.

Second, fines give a fairly good idea of where money laundering is taking place and how serious it is.

By amount the bulk of money laundering fines are levied by regulators in the USA and Europe on banks operating within their jurisdictions. The size of these fines indicates the volume and seriousness of the infractions.

According to AccountancyDaily, there were 58 AML penalties worldwide in 2019 totaling USD 8.14 billion nearly double the amount of the 29 penalties in 2018.

US regulators were the most active with 25 penalties totaling USD 2.3 billion, the UK with 12 penalties totaling USD 388 million.

(Fines like the UK's aren’t going to do much to combat money laundering. They’re unlikely to silence the laughter in board rooms.) 

France took the record that year for the largest penalty USD 5.1 billion levied against UBS who are appealing the fine. 

In terms of money laundering and sanctions violations penalties (the latter a particular preserve of the USA) there have been some USD 36 billion since the 2008 financial crisis according to a recent report by Ferengo who publish this data annually. 

Again the countries whose banks were “tagged” with the fines come from “developed and well-regulated” markets. You won’t find Angola or Iran among them.

Third, despite apparent precision attempts to quantify money laundering, corruption, and other illicit activity results in crude estimates at best and often in semi-educated guesses. Sometimes they measure less than we think they do.

Here are some examples from previous posts.

Sometimes they don't model everything we think they do.

For example, Transparency International’s Corruptions Perceptions Index is based on perceptionsThere are no public statistics on corrupt transactions, except those arising from legal cases.

More importantly TI considers only governmental corruption as clearly outlined on TI's website.

A fuller discussion here

The latest TI CPI rated Denmark as the least corrupt country in the world tied with New Zealand.  In 2018 Denmark stood atop the CPI alone.

Yet, Den Danske Bank’s branch in Estonia was involved in a Euro 200 billion money laundering scandal over more than a few years.

According to reports, the Central Bank of Russia informed both Danish and Estonian regulators in 2007 and 2013 that Danske's Estonian branch was being used for money laundering.

It would appear that warning was not acted upon, presuming that these reports are correct.  

Denmark and several other TI highly rated countries are victims in a multi billion euro tax fraud based on “dividend stripping” or a “cum/ex” operation. Denmark Euro 2 billion, Germany 5 billion. 

Major European banks colluded with clients to implement the scheme. Not one Angolan bank in the mix.

So, if you’re relying on TI to frame your money laundering risk profiles, think again. It's only a partial source and has its own limitations. 

Most money laundering takes place through private sector not public sector banks. And be aware that other rankings of financial crimes risk use the CPI as input. 

Fourth, assessments (mutual evaluation reviews) of anti-money laundering regulations and systems miss key factors.

Much of these are based on whether there are laws on the books that cover the key elements in a good AML regime (based on FATF recommendations) and how strict they are.

Geography plays in important role in this process because "zip code" is used to risk exposures.

MENA countries are likely to be scrutinized more for terrorist finance than Latin American ones. Deficiencies found in MENA will be treated as more serious than those in Latin America.

Checking the robustness of legal regimes is certainly important. 

But, if laws are robust but ignored, they are of little use.  If they provide less than perfect solutions as all legal systems do, then reliance on them should be tempered by this knowledge.   

The same with individual financial institutions procedures for enforcing the national AML regime.

Fifth, compliance with AML regulations involves a lot of "box ticking".

Financial institutions are required to perform two key AML tasks:  (a) conduct CDD (customer due diligence) prior to accepting a customer and (b) monitor a customer's transactions and conduct of its relationship to detect suspicious activity. The bank should investigate suspicious activity and where warranted file a report with its financial regulator.

CDD consists of "proving" a customer's identity, financial condition, etc. and assessing whether the potential customer is not engaged in illegal or terrorist activity. 

This involves obtaining various documents depending on whether the potential customer is a legal entity or a natural person, e.g. passports, commercial registrations, financial statements, etc. 

As a general rule, one can say that individuals and businesses that need their money laundered have rather high gross margins. Thus, they have ample budgets to pay intermediary fees and make facilitating payments.

They have the money to pay for the creation of documents to comply with AML requirements and to set up complex structures to hide realities. 

Here it's important to note banks are not required to “verify” documents but can accept them on their face as valid unless the documents are clearly forgeries or inconsistent. This is similar to their obligations regarding the authenticity of documents presented under letters of credit. 

That makes sense.

Unless it’s a rank amateur forgery how will a bank in Country A, verify your Country B passport?  Or your financial statement?  Or a letter from a lawyer confirming you inherited USD 100 million from your Uncle Abdullah?

Money launders are also able to hire “name” professional advisors and intermediaries. 

The sort of “names” that would add a “halo” to the client.

"If XYZ is dealing with Ms. X, she must be “clean” because they are an international firm and must have done proper 'due diligence'.

Most national AML laws and regulations are based upon the recommendation of the FATF an international body that develops AML standards. FATF Recommendation 17 allows a national regulator to permit its banks to rely on a third party to perform some of the key steps in CDD or customer vetting.  

This can also be a convenient excuse for a bank not to look too closely for fear of finding out something it doesn't want to know or aggravating a potentially lucrative customer.. 

What's hopefully clear from all of this is that there is a great deal of "box ticking" in the CDD process. 

As an internationally mobile individual, I have had to open accounts in a variety of foreign jurisdictions.

In most cases I have been called on to sign a letter that states I won’t use the account for illegal purposes or to fund terrorist activities.

I also have been asked to self-report my income and net worth. Provide proof of my address, e.g., a letter addressed to that address, a utility bill, a driver's license, etc.

But not in every case!

And that's not just overseas, in one case when I lived overseas and wanted to open an account with one of the largest banks in the USA, my Social Security number, a deposit, and a smile were all I needed to open an account.  I wasn't asked to "show" my passport or driver's license! 

Luckily for the banks I deal with, I’m an honest person. But not every client is.

This concept of self-certifying oneself as “not a crook” is an interesting one.

Once while opening an account with a foreign bank, I asked its representative if he thought that a person who wanted to use an account for illegal purposes or terrorism would have his or her conscience troubled by lying in such a letter. 

His retort was “If you don’t sign the letter, no account”. 

Letter signed, box ticked, financial integrity protected, we moved on.

To be fair, how would a bank determine if a potential customer were a terrorist or criminal?

Are banks smarter than law enforcement and certain organs of state security?

Do banks have access to information that these official bodies do not?

Bank financial performance seems to suggest that as a group they are less "smart" and have less access to information about their clients than is commonly imagined. 

Once a customer is accepted then the bank must monitor the account relationship for transactions that don't seem to be compatible with the information provided on financial capacity or contain money laundering patterns. The FATF publishes a list of these, if you're interested.

In general this is "easier" than the CDD step.  A lot of this can be automated fairly simply primarily where the account is being used for many transactions.  

However,if the account is being used for a one off or infrequent transactions, this becomes a harder task.  

Six, The phrase "pecunia non olet" explains why money laundering occurs.

In fact the more you have the sweeter the smell, particularly when fees for advising and handling that money are based on amount.

Ms. dos Santos was clearly a “PEP” (politically exposed person). 

Standard practice AML procedures would require that she be subject to enhanced due diligence at the inception of a relationship and thereafter by ongoing enhanced monitoring by banks, professional service providers, and intermediaries.

She must have had quite a convincing story about how she legally accumulated her wealth and then kept on legally accumulating more. 

Plus a trove of very credible looking documents.

Or maybe it was the “sweet smell” of success?

Or the fear that too intrusive an approach would take her to finding another bank, advisor or professional service provider? 

The rich are different than you or me.  Not only do they have more money than we do.  But their bankers are more deferential.  

Perhaps, a even more compelling story is that of Rabobank NA, the US subsidiary bank of Rabobank in the Netherlands.

According to the US Department of Justice, two of the bank's rural branches in California--Tecate and Calexico cities right on the border with Mexico--had a "lot" of US dollar deposits.

So much that the bank had to send at least one armoured car per week to collect the cash. Depositors then typically wired or otherwise transferred the money out of the accounts shortly after making the deposits.

Now one (at least AA) would think that large cash deposits well beyond economic activity in the area--at least legitimate activity--might be a "red" flag of potential money laundering, warranting at least an investigation.

The immediate transfers would typically also be considered a "suspicious" transaction.

Well Rabo did investigate and apparently decided it was.

So they did what any "responsible" financial institution might do.

They devised a "verified customer scheme" to facilitate continued acceptance of the deposits in an ultimately unsuccessful attempt to hide the transactions from USA authorities. A seemingly high risk strategy.  What were they thinking?. 

Thursday, 13 February 2020

Goldilocks and the 3 Bears - A Financial Fairy Tale - Part III Dana Gas

Almost Golden Once More
Links to earlier posts in this series Post 1 and Post 2 

On 20 June 2017 Dana Gas announced that Goldilocks’ shareholding had reached 5% based on information it received from the ADX.

At the same time local press reported Goldilocks’ announcement that it had purchased 350 million shares in Dana Gas giving it a 5% stake.

Looking over trading on the DFM, there was a spike in volumes concentrated in the period 5 June through 18 June.

Why stop at 18 June?

As per its regulations, the ADX had to notify Dana when trades caused a shareholder’s shareholding to reach 5%. Presumably this notice to GFH came late on 18 June or 19 June. Dana then had to prepare its announcement and send it in. 

Also trading in Dana shares was relatively light on 19 June..

What we do know is that Goldilocks reached the 5% reporting threshold sometime around 18 June.

But we don’t know if this is because it bought 5% during this period.

Or if it held a position previously and only a modest amount of purchases took it over the threshold.

I think it’s likely that Goldilocks purchased the 5% around this period. This would be in line with their “constructionist” philosophy, i.e., an active investor. They don’t seem to be the type of investor who would accumulate a passive position.

But who knows.

If you remember this is during the early stages of DG’s restructuring of its previously restructured sukuk. So there was a “dip” in price.

What better time for a savvy investor like Goldilocks to buy?

Some details on DG stock transactions.

From 5 June to 18 June the volume of trades in DG stock increased dramatically. Some 84% of trading volume between 1 May 2017 and 18 June 2017 incurred during this period.

For the full period 5 June to 18 June, 1,548,351,688 DG shares were traded with a total value of AED 921,047,016.63. Average price AED 0.60.

For the period 12 June to 18 June, 1,020,149,231 DG shares were traded with a total value of AED 698,149,961.05. Average price AED 0.68.

For the period 14 June to 18 June 632,217,208.00 DG shares were trade with total value of AED 449,382,287.83. Average price AED 0.71.

Note the above are value weighted averages based on volume and value of shares traded not closing prices only. However, the closing price averages are almost the same.

That gives us three plausible entry costs: AED 0.60, AED 0.68 and AED 0.71 per share.

One wrinkle.

According to page 181 of the Listing Particulars for the Nile Delta Sukuk as of 31 December 2017, Goldilocks owned 4.68% of DG’s shares.

I have not been able to find an explanation for the decline from 5%.

Selling 0.32% of its holdings seems a strange thing to do, particularly because just a short while later it increased its holdings to 5.21%.

Since Goldilocks announced its 5% holding, it would seem that this wasn’t a maneuver to disguise its status as a “major” shareholder.

For those with an interest in which "fish" may be in the same "school", among other shareholders mentioned in the Listing Particulars are Ajman Bank, Bank of Sharjah, Shuaa etc. 

Whether other members of the ADFG/IC are the UBOs of these shares or not is not determinable from the information at hand (at least AA’s hand).

On 6 February 2018 The National reported that Goldilocks had increased its share to 5.21% and that ADFG/IC owned roughly another 3.79% for a total of 9% of the company.

Goldilocks announcement was made the day prior.

I’m going to ignore the information in the Nile Delta Sukuk Listing Particulars and assume that prior to this increase Goldilocks held 5%.

As FYE 2017, DG had 6,976,623,442 shares of stock. 0.21% is equivalent to 14,650,909 shares.

On 6 February the day prior to the announcement, 61,978,999 DG shares were traded for AED 44,739,016.70. Average price AED 0.72.

If we assume—and AA does—Goldilocks reported shortly after buying more shares, then this is a likely day for the purchase.

Another wrinkle.

As of today, according to the ADX, Ajman Bank holds 5.59% of DG shares.

Did Goldilocks sell it shares to Ajman Bank or a customer of the bank in a “true” sale?

Or did it “sell” the shares to Ajman to “perfect” Ajman’s collateral for a loan to finance the shares?

If Goldilocks is still the “owner” of the shares, then it appears there is a less rigorous concept of major shareholders and related party groups than for example on the Bahrain Stock Exchange

There the UBOs for GFH shares registered in Bank AlHilal’s name are disclosed. On the DFM they are not.

It would appear, and I hope I’m wrong, that in the UAE if the shares are registered in the Name of X, never mind if Y is the UBO. If X is related to Y and Z, never mind considering them a group.

So how did Goldilocks make out on this investment?

The market price of a share of DG stock on 13 February 2020 was AED 0.92.

In 2018 and 2019 it received cash dividends of AED 0.05 and 0.055 per share.

Ignoring the time value of money that means the exit price is AED 1.025.

Based on the most favorable entry price of AED 0.60, Goldilocks had a 71% return over the period. Note that is not a per annum return.

Based on AED 0.68 entry price, the return is 51%.

Based on an AED 0.71 entry price, the return is 44%.

Note: I have not adjusted the entry price for the February 2018 purchase of 0.21% in additional shares.

That’s because we’re working with estimates. There’s really no point trying to add precision where there are so many assumptions.

Cash on cash AED 0.425 gain per per share equals AED 149 million (USD 41 million) on Goldilock’s 5.21% stake.

To sum up: GFH a loss of USD 23 million; Noble a loss of USD 38 million, DG a gain of USD 41 million. A net loss of USD 20 million over the lives of the investments.

But let’s turn to a year by year allocation to see the impact on FY2018.

As you’ll recall from Post 1 and Post 2, for FY 2108 Goldilocks had an estimated loss on GFH and Noble of between USD 55 million to USD 60 million, depending on its FYE 2018 valuation of its Noble Group Holdings shares.

As per the table below, Goldilocks estimated FY 2018 earnings on DG are AED 70 million (roughly USD 20 million). So for FY 2018, the 3 Bears represented a loss of USD 35 to USD 40 million, again depending on the FYE value of Noble.

Goldilocks Investment Returns in Dana Gas
Millions of UAE Dirhams (except share price)

Date Price MTM Cash Stock TOTAL
20Jun17 0.60

31Dec17 0.81 73.50 0.00 0.00 73.50
31Dec18 0.96 52.50 17.50 0.00 70.00
31Dec19 0.96 -0.35 19.25 0.00 18.90
13Feb20 0.92 -13.65 0.00 0.00 -13.65

112.00 36.75 0.00 148.75