A-Mark Precious Metals (AMRK): A Dealer and Lender in Physical Gold
A-Mark Precious Metals (AMRK) has been written up twice at Value Investor’s Club. The most recent time was this year. You can read those write-ups over there. It was this most recent write-up at Value Investor’s Club that got me interested in the stock. However, it was for different reasons than that write-up itself lays out as the case for buying the stock. The VIC write-up focuses on how low volatility in the price of gold (and silver and other precious metals) in recent years means that A-Mark has under-earned in each of the last 5 years or so. Having looked at the company now – I’d say that’s possibly true. A-Mark says many times in its SEC filings that it benefits from increased volatility in the physical markets for precious metals. The company also says that the price of gold – rather than how much that price bounces around – doesn’t much matter to the company’s results. I’m less sure of this second point. There is one activity that the company engages in where I feel high (and continually rising) gold prices would be a benefit and low (and continually falling) gold prices would harm the company. Since I mentioned “activities” – let’s talk about what acts A-Mark actually engages in.
The best way I can describe this company is as an investment bank (really, a “trading house”) focused on physical precious metals. That word “physical” is important. We are talking about the buying, selling, storing, shipping, minting, lending, and many other things of actual physical bars and coins of gold, silver, etc. The business is almost completely U.S. It seems 90% of profits probably come from the U.S. I say “seems” and “probably” because of some difficulty in using traditional accounting measures when looking at a company like this. A-Mark is a financial company. It really is a highly leveraged and fully hedged – or as near as fully hedged as it can get – trader in a market. As a result, an accounting line like “revenue” is meaningless. The company reports revenue. But, revenue doesn’t matter. The first line on the income statement that is worth paying attention to is “gross profit”. Gross profit at A-Mark is always less than 1% of revenue. Usually it’s quite a bit less than 1%. This makes typical SEC requirements to disclose revenue stuff useless. For example, does A-Mark have high customer concentration? We don’t know from the 10-K, 10-Q, etc. There’s a line in the 10-Q that says about 50% of revenue comes from two companies: HSBC and Mitsubishi. However, this is just hedging activity. Because of how A-Mark’s accounting works, you could list big “customers” as just entities they are making sales to in the form of hedging activity that will never be settled with physical gold and will never result in any gross cash profits for A-Mark (on their own). So, it may be that around half of the revenue line you are seeing is hedging done with HSBC and Mitsubishi. But, that doesn’t help us understand A-Mark’s business. The amount of sales they make – and, remember, I think these particular sales to those big banks would never result in physical delivery of any gold or other metals – are just the end result of how much business they are doing with their actual customers. We’re seeing the other – hedging – side of their real business. This bit about HSBC and Mitsubishi isn’t important. It’s just a digression I’ve included here to make very clear that you can’t rely on comparability between A-Mark and any other public company financials you’ve ever seen. You can’t go to a site like quickfs.net and look at the last 5 years of financial results and think you really know what happened with revenue, margins, returns, etc. It’s a lot more complicated than that. And you have to break the company down yourself.
I think A-Mark can be broken into 3 parts. One part is “GoldLine”. This is probably the part anyone reading this is most likely to be familiar with. GoldLine has historically run a lot of cheap TV ads, radio ads, etc. trying to sell gold directly to American households. The business was supposedly EBITDA positive almost all the time before being bought by A-Mark. It’s currently losing money. The consolidated results you see for A-Mark include losses at GoldLine. The business was acquired recently by A-Mark. It may be disguising some of the earning power of A-Mark overall. I generally like to see stocks where there is one business segment losing money. This sounds counterintuitive. But, it’s usually the easiest thing a company can fix. If you stop putting money into the loss making business, you sell it, you liquidate it, or you turn it around – profits suddenly jump. A lot of investors screen for low P/E, low EV/EBITDA, etc. and for high ROIC, ROE, margins, etc. Well, screeners combine all business segments in a corporation when showing you results. So, a stock with 2 great businesses and one terrible business can sometimes keep it off screens. I had success buying into Babcock & Wilcox – which later became BWX Technologies – when it had one great business, one mediocre business, and one speculative money losing start-up. The company shut down the start-up and broke the rest of the company in half. The result was a very rapidly rising share price at the great business. Very little changed. It was really the same business before and after. But, people actually valued the great business alone higher than they had been valuing all the parts together. Is A-Mark going to break up? No. But, it has already started slashing operating expenses at GoldLine. We’ll see if GoldLine was a dumb acquisition or not. But, overall – a lot of what the company has been doing in terms of what they’ve acquired does make a lot of strategic sense to me. Whether or not it was done at good prices is hard for me to tell at this point. But, the company does seem to stick very, very much in the center of its “circle of competence”. Everything it does is focused in some way on being a dealer in physical precious metals.
So, GoldLine is the money losing business. What are the 2 money making businesses? One is what I’ll call the trading business and the other is what I’ll call the lending business. A-Mark calls them “trading and wholesale” and “secured lending” (or something like that). For this part, I’m not going to go into a lot of detail about what I personally would pay for A-Mark. I’m just going to trty to take a stab at what the company might be valued at in a more general sense. Given today’s corporate tax rates – up to 79% of pre-tax income can convert into after-tax profit at a U.S. corporation – I’m going to assume that a business could be worth 12 times pre-tax profits. This is equivalent to a P/E a bit over 15. Of course, a company could deserve a P/E as low as like 10-12 or as high as 30-35 depending on other factors like what the return on shareholder’s money is, whether it’s growing, etc. For me, these are difficult to calculate for A-Mark right away. Knowing what normal leverage and normal profitability is – given the fact that volatility in precious metals markets influences the company’s profitability – would be tough for me to figure out this early in the research process. So, I’ll start by just describing these businesses as best as I can understand them and then slapping some very rough multiples on them and seeing if that is anywhere near the current stock price.
Okay. So, secured lending. This is the business that interested me the most when I read about it. I compare it to margin loans on physical gold. The company has not had any loan losses in this business segment, and – given how they describe the loans – I don’t see any reason why A-Mark should ever have meaningful loan losses in this business absent something like fraud, employee misconduct, etc. These loans yielded 9-10% in the last year or so. You can see why I focused my attention on this segment first. A business that can lend at 9-10% with theoretically close to zero risk of loan losses should be an attractive business.
A-Mark buys (60% of the loans in this segment were acquired from other lenders) or originates (40% of loans were made by A-Mark itself) short-term loans backed by gold bars and gold coins. These loans are made at a loan-to-value ratio that’s low compared to the marketability of the collateral (hence the lack of loan losses). The company often makes loans that amount to 50% to 85% of the value of the gold. A-Mark claims it is using the liquidation value of the collateral when making this calculation. If so, it’s possible that the loan-to-“retail” value for the coins is quite a bit lower than the bars. A-Mark has physical custody of all the collateral. Margin calls usually occur at 85% loan-to-value ratios. These loans vary in length from 3 months to 1 year and vary in size from $15,000 to $10 million. If you look at the results A-Mark puts out about growth in various business segments – it’s clear that a lot of these loans end up getting closed out through liquidation of the underlying collateral. There are a lot of margin calls. This is the segment I think may be more tied to higher and rising gold prices than A-Mark claims. As gold prices fall – a lot of collateral calls occur, A-Mark closes out more loans than clients open, etc. As gold prices rise, there is more and more collateral value to borrow against. Basically, the borrowing capacity of A-Mark clients should widen as gold prices rise and shrink as gold prices fall. This is the one long-term aspect of A-Mark’s business that seems unhedged to me. My thinking right now is that one business segment – trading – is influenced by volatility. Another business segment – lending – is influenced by gold prices and nominal interest rates. Very high nominal interest rates, very high inflation, very high gold prices, etc. would probably be good for the “secured lending” business at A-Mark. The reverse – low nominal interest rates, deflation, low gold prices, etc. – would be bad. A-Mark has about 7 million shares outstanding. Recently, “lending” has made about $2.2 million or so on average in net interest income. If we capitalize that at 12 times, we’d get $26.4 million in value. Divide $26 million in value by 7 million shares and you have $3.70 a share in value from lending. Even if we said – okay – it might be worth $4 a share, we have a problem. The stock is trading at $11 a share. If the business segment I like most is only worth $4 – that means at least two-thirds of the business value (just to get me a 0% margin of safety) is in a business segment I have less confidence in.
That business is “trading”. This is the part of the business that might benefit a lot from higher volatility. I’ve now read a lot about this segment. I know what it does. But, I don’t know how to decide what “normal” earning power is here. If I had an extremely long record of financial data for the business unit I could compare it to volatility in gold prices in any given year and come up with estimates. It’s easy to get gold price data for all years. So, I can easily calculate normal gold price volatility myself. But, I can’t correlate it with A-Mark’s trading unit profitability without a longer record. For that reason, I’m just going to take the most optimistic look at the last 2 years of results. If I exclude interest income and interest expense as well as “other” items and just take the last 2 years of gross profit less the last 2 years of SG&A and then average them – I get a figure of about $6 million per year. That’s probably an overly optimistic estimate of the earning power of this trading unit in years like 2017 and 2018. There is enough data to calculate EBITDA for this business unit – but, I’m not sure the D&A should be fully excluded. Cap-ex is similar to “D&A”. For that reason, I’m just going to say the operating profit at this segment was about $6 million – on average – over the past 2 years. These were non-volatile years for gold. So, perhaps “normal” earning power is a lot higher. The per share amount that translates into is $6 million * 12 = $72 million. We then divide by 7 million shares and get $10.28 a share. We got something like $3.70 for the lending business. So, we add them up and get around $14 a share for an $11 stock. Not much of a margin of safety. And, unfortunately – 2/3rds of the business value is in the segment I understand least.
There’s a long corporate history here, a spin-off, a past of some bad corporate governance, etc. I’m going to skip all that. At this point, A-Mark is a pass for me. It looks fine as a speculative bet. I tend not to make those. It isn’t correlated at all with stocks generally. If someone really wanted to own something gold related – maybe this is what they should own instead of the metal itself. I don’t know. It’s not obviously expensive in any way. It seems cheap enough based on earnings that are probably lower than average. I don’t have evidence one way or the other of whether it’s especially safe or dangerous. So, I can’t bring myself to give it my lowest rating. But, it’s definitely a pass. I’m not going to re-visit this one.
Geoff’s Initial interest: 20%