I have written book reviews in the past in bits and pieces. However, it has been a long time wish to make a comprehensive list of the best investing books that I have read. Such a list goes beyond the typical finance focus and takes a multidisciplinary approach, which is something investment legend Charlie Munger had been advocating for. Hence, in this long (really really long!) piece, not only you will find the usual topics such as valuation and accounting, but also life advice, culture, philosophy etc. The list is however only limited to books that I have personally read and so don’t be disappointed if one of your favorites did not make it. I know that some great books are missing and once I have read those I will include them also.
Tracking interest rates in Bangladesh is no easy task!
Banking sector developments have dominated newspaper reports for the last few years. Although non-performing loan formation took the spotlight for a long period, since late 2017 the soaring rates on fixed deposits and tight liquidity conditions also came under attention. Much has been written on why the interest rate cycle has turned so sharply and so I move to a slightly different territory.
My interest however in on the public data available on interest rates. Generally we can follow a number of the indicators from the central bank website. One is the call money rate, the rate at which one bank lends to another. A quick glance at the rates will show that despite all the hue and cry about liquidity crisis the call money rate has barely bulged. This has happened because there seems to be a verbal guidance from the central bank on what the call money rate should be. This obviously makes the call money market less liquid and prevents it from being an indicator we can use to gauge interest rate conditions.
The second option is to look at yields on government bonds. There also we have a problem as the government has ceased to borrow from the bond market for more than 2 years and instead has been exclusively relying on National Savings Certificates. This lack of demand from the government depresses bond yields and prevents it from showing true liquidity scenario. The 10 year government bond yield is of particular interest to equity analysts because the Capital Asset Pricing Model uses it to form Cost of Equity in valuation models. The 10 year government bond yield in the January auction was 7.39% which means it has barely moved from the lows reached. Using this rate in valuation models will inevitably lead to asset overvaluation.
A third option to track interest rate is looking at weighted average deposit rates. This data comes with a bit of lag and because the rate is an average not just fixed deposits but also current and saving deposits it will not capture the size of the move in interest rates.
Regardless of these data challenges it is very clear that interest rates in reality has gone up. Banks which not too long ago offered as low as 5 to 6% on fixed deposit have raised that to 9 to 10%. A 400bps move in a matter of months is a very steep move.
Interest rate and stock prices. What does the theory say?
In a discounted cash flow (DCF) framework keeping all other things constant a higher interest rate in the economy will lead to lower values. The transmission mechanism is generally via the risk free rate (aka 10 year government bond yield) which goes up with interest rates and we end up with a higher cost of equity. Discounting future cash flows with higher discount rates will lead to lower values than the past.
There is a common misconception about companies which have a lot of net cash balance in their balance sheet. We can take the example of Square Pharmaceuticals where cash and marketable securities make up 38% of assets (there is barely any debt either). From an accounting perspective, higher interest rates would result in higher interest income on their cash balances and higher earnings. However if we use the DCF framework then the impact of higher interest rates is a net negative.
Value = Net Cash + PV of future cash flows
Net Cash = Cash – Debt
The net cash on balance sheet is today’s value and is not sensitive to interest rates. The Present value of future cash flows on the other hand reduces leading to a lower value. What is however true is that companies with 0 net cash balances are more sensitive to interest rate hikes and will suffer more relatively. So a company like Square with lots of cash in hand will be less sensitive than companies burdened with a lot of debt like ACI.
What if we don’t keep all other things constant?
So far in our discussion we only changed interest rates but made no reference to cash flows. Small changes in interest rates may have no effect on the real economy. But large changes can lead to reduction in consumption as aggregate demand comes down. So some businesses might feel a double whammy as future cash flows could decline and the discount rate used on those flows could go up.
We also need to be clear about what is causing interest rates to move up. An increase in long-term government bond yield could happen because of expectations of faster economic growth (this increases the real risk free rate). In this case however, we are actually revising up our estimates of future cash flow but using a higher discount rate. The net result could be either positive or negative.
Does real world data confirm the theory?
Reality is always a bit more complex and messier than theory suggests. By and large periods of higher interest rates result in lower stock prices (and the theory is proved). However, there are periods in history when higher interest rates coincides with increasing stock prices. In the chart below in the period between 2004 and 2007 higher stock market and higher interest rates happened at the same time.
There are many explanations for these periods of contradiction. A simple one could be that higher rates were accompanied by heavy economic growth and thus cancelled each other. Another could simply be that market prices had already incorporated higher interest rates and thus when rates actually went up asset prices did not come down.
Remember our valuation methods provide estimates of fair value and market prices at times could be substantially below the fair value before rates have moved up.
Let us have a quick recap of what we have discussed so far.
1. Despite no clear sign from the popular data on interest rate, interest rates in the financial sector have moved up a lot.
2. According to finance theory, higher interest rates should lead to reduction in equity values. Stocks with net cash will also be negatively affected but may perform better than those with lot of debt.
3. Cause of higher rates are also important. Higher economic growth can lead to higher long-term bond yields. This may have a net positive effect as we revise our cash flow projections upwards.
4. In practice there are periods of anomalies where higher interest rates coincided with higher stock prices.
The real effect of interest rates on stocks therefore depends on prevailing stock valuations. If rate hikes are happening when stock valuations (computed using P/E or P/B) are already very high then we will inevitably see stock prices decline. If they aren’t then its harder to predict how prices will end up reacting.
Source: Data for the first chart is from Bangladesh Bank. The bond yields only show yields on new auctions. The data from the second chart sourced from a Quora question on the same topic.
I have agonized over the role of cash in stock portfolios for long. Often I felt that I should have some cash at all situations because market throws some very attractive opportunities from time to time. Other times I felt that by not remaining fully invested in the stock market I am trying to time the market which is impossible. Generally I have jumped around from one school of thought to another. Before I go into my latest thoughts on cash let me share what two popular investors practice.
One of the first thing we learn about investing is the price to earnings ratio. Analysts spend a great deal of time trying to forecast earnings and whether companies meet analyst expectations of earnings tend to move stock prices in the short-term. What is however surprising is that valuation models of companies use cash flows and not earnings. Thus, its cash flows and not earnings that determine the value of companies.
By reading about many investors I have realized that looking at an investment case from different angles and perspectives unearth insights which can be very valuable. For example, in a simple DCF framework we value a company only based on its long term cash flows. However, the assets of the company could be used for something totally different and unrelated which can create more value.
There was a story I read about a chain restaurant which was struggling only to be eventually taken over by a smart guy. He sold some of the locations and rented out some other as office space etc. The assets therefore were acquired by the guy who could bring out the maximum value in them. I came across an article in the blog Basehitinvesting which to me was a very interesting way of looking at a company. Instead of trying to find undervalued companies we could try to identify undervalued products and back calculate to see if the company looks undervalued if pricing of the product is maximized.
US equities are trading at a P/E ratio of 25x and a Shiller P/E (aka CAPE) of 27x. Since 1870, the mean P/E ratio had been 15.6x while the mean Shiller P/E is 16.7x. A somewhat naïve conclusion is that the market is overvalued as it trades above its mean and if financial markets have any chance of mean reversion equity holders will lose out big time. There are however several problems to this conclusion.
- It does not bring into consideration the extremely low interest rate environment. The current 10 year government bond is yielding 1.6% while the average since 1871 is 4.6%. Intuitively we can understand the inverse relationship between P/E multiples and interest rates. This article sheds some color on the topic. It is only natural that stock valuations will remain high in this environment.
- Selling stocks will mean the cash has to be deployed elsewhere. The problem is that the low interest rate environment has resulted in high prices for other asset classes including fixed income and real estate. This caused a few commentators to comment that staying in overvalued equities might be better than shifting to overvalued fixed income and other asset classes.
- A P/E higher than mean does not always lead to poor returns in later years. In fact on many occasions market has performed even from a high P/E level. However, when the P/E ratio goes way out of line it does indeed result in a poor 1 year or 3 year returns. Thus some investors prefer to remain fully invested at all times. Also, there have been periods when modest rises in interest rates have not caused stock under-performance as they coincided with an acceleration in economic activity.
- The mean P/E we calculate can be affected significantly by the time frame we choose. There is no scientific method of choosing the perfect time frame.
One thing is however clear. If interest rates start to mean revert (if it goes up a lot), value of all asset classes will suffer. This is why people are so concerned about interest rate decisions by the US Fed. Yesterday (22nd September), Fed kept interest rates steady but indicated that there could be a hike by the end of the year. Yet trying to predict the path Fed will take is a gargantuan task.
While reading books and listening to podcasts I came across a variety of different such methods people have employed to deal with such an environment. Each of these strategies have their own pros and cons. There are also limitations that might prevent portfolio managers from using some of these. For example, a pure equity fund manager might have limit on how much cash he can keep. Similarly someone mandated to invest in US equities cannot move to other geographies or other asset classes.
- Actively look for undervalued assets: The best strategy would be to look for companies which are trading far below their intrinsic value. This gives the margin of safety that could limit downside risks. The best example that I read was something Marathon Asset Management described in their book Capital Returns. In an environment when equities became expensive they choose companies with pricing power and strong economic moats. Their portfolio traded at a similar P/E ratio similar to the broader market with one major difference. The FCF/Net Profit ratio of their portfolio was 90% compared to 50% for the market. On a free cash flow metric the portfolio was actually cheaper than broader market. The strategy appealed to me because in an overvalued market all assets will typically look expensive and it will be extremely difficult to pick stocks which look cheap on traditional multiples. However the smart analyst can find companies which ‘appear’ to be expensive but in reality are not. In Marathon’s case they prioritized cash flow capabilities. Other examples could be companies which are making investments for the future but maybe the accounting treatment makes them ‘expense’ these investments instead of capitalizing.
- Move into cyclicals with better risk reward: This strategy is in one way almost the exact opposite of the first method. This is the strategy that Simon Hallet of Harding Loevnor explained in an interview with Consuelo Mack. In order to buy companies at reasonable valuations they moved into more cyclical companies which as they explained could have volatility in the near term (which they are willing to tolerate) but the underlying economics of these companies should allow them to perform in a longer time horizon. While Simon did not elaborate on which sectors they invested in one can guess that he would avoid cyclical sectors that are at high valuations and may have invested in those that have come off a lot. An example could be commodities which have been hurt by lower demand from China and in case of oil higher supply. The reason it is the opposite method to Marathon’s is because cyclicals are unlikely to have economic moats or pricing power. However, looking from another perspective Harding’s approach is actually similar to Marathon’s. At the end of the day both are trying to buy equities below their ‘intrinsic value’ (which is in its true sense value investing).
- Diversify abroad: For US investors one option that has been described by many experts has been to diversify to overseas equities. Emerging markets (EM) has been mentioned as EM equities had been under-performing US equities for a long period. It is true that many of the larger EM markets like Russia, China and Brazil have their own macro issues. Yet that should not prevent anyone from at least searching for ideas.
- Move into different asset class: Another strategy that I heard several times is owning some gold. This has less to do with higher interest rates and more to do with the possible aftermath of the massive and unprecedented monetary expansion done by central banks around the world. A few investors also have no problem holding cash even though they could underperform the broader indices with that approach.
- Form a market neutral portfolio: In layman terms if we are long on certain stocks and short on some others, the market direction might not matter that much as we should theoretically be able to eliminate the beta. A somewhat different and more advanced method is the one used by Bridgewater Associates even though they are more a macro fund than a bottom up stock picker. By studying long periods of macroeconomic data they have built internal models of how certain asset classes behave in certain environments. They use these conclusions to create a number of potential trades at any given time. These trades minimize correlations and hence the term ‘Pure Alpha’.
Finally, sometimes the world is such that we need to be prepared to have periods of low returns. The focus at such times should be to lower permanent capital losses. By protecting the portfolio one can set up a stage for decent returns in the future.
Cyclical stocks are those that have high correlation with economic activity and thus see a lot more volatility in revenue and earnings. Typical examples will be banking, cement, metals & mining etc. However, one needs to be very careful when categorizing companies into cyclical or non-cyclical. Not all banks need to be cyclical because sometimes lack of private sector credit growth can be offset by financing the government. Similarly companies which we would generally consider displaying secular growth characteristics might turn out to be more cyclical.
Screening for cyclical companies are a bit trickier than others. This is because simple extrapolation of the trends will often lead to over or underestimation of fair value. In addition, at the bottom of the cycle earnings might be exceedingly low (as some costs remain fixed while sales drops) resulting in an extremely high Price/Earnings ratio which may cause people to unfairly tag the company as overvalued. To deal with these issues here are some tips on the subject.
- Use a sales multiple for screening: Instead of the more commonly used P/E ratio it makes a lot more sense to use a Price/Sales or better yet an Enterprise Value/Sales ratio. This will help negate the impact of operation leverage that causes profit to drop sharply in an economic downturn.
- Watch out for replacement value: One good strategy is to compare enterprise value of listed equities with their replacement values. When they are trading very near or below their replacement value, it makes more sense for a potential acquirer to buy out the listed company rather than make a new factory.
- Leverage is an useful indicator: When the entire sector becomes highly leveraged it is a good bet that capacity additions will become slower and eventually demand-supply parity will come back in the industry. Also, in a cyclical downturn, companies with strong balance sheets can become opportunistic and acquire over-leveraged competitors at rock bottom prices.
- Some companies can be defensive to the cycle: Integrated oil companies like Exxon Mobil might not be as hurt as those which only have Exploration & Production business. That is because, in a oil price decline while the E&P segments will perform poorly the oil refinery business can benefit from higher margins (the drop in retail oil prices has been lower than the drop in crude oil prices resulting in thicker margins for refiners).
These are few methods that can be put to use by equity analysts. Obviously screening is just the first stage of idea generation and one needs to do further analysis and due diligence before making the investing decision.
A country saw weak loan growth of 7% on average for around 10 years. Every year people expected growth to pick up only to become disappointed. Then in the beginning of 2016, two analyst’s predicted that loan growth will be 15% which is double the average for the previous decade. At the end of 2015, real growth came exactly similar. Both analysts were correct. Yet, for the purpose of investment management the value of the two forecasts can be starkly different.
Why is this? For the purpose of investment management we need to take action before the event has taken place and potentially before Mr Market has figured this out. For a buy side PM to make this decision he needs to be convinced that this time is indeed different. The analyst who simply projected growth to pick up without backing it up with logic will not help the PM at all. On the other hand a carefully crafted analysis, which explains in great detail the underlying drivers of loan growth, barriers to growth that were present for so long and what has changed will be immensely valuable and most importantly actionable.
There are many types of value added work that can be done to make the case.
1. Combing through the breakdown of loans to see which segments made up bigger piece of the pie. If these segments are not growing it will be hard to move the needle.
2. Checking if any small segments have been growing rapidly and despite being a small part overall may become important enough in influencing the total number.
3. Finding which segments are underachieving and finding the reason for such poor performance. Quite often it could be due to energy and infrastructure bottlenecks. Sometimes industries can go through structural declines as well in which case credit growth might not recover.
4. Understanding the business cycle and interest rate cycle to see if they can explain the underperformance of credit growth.
5. Checking if companies are being able to get alternate sources of financing from international institutions or non-bank finance companies which has eaten into bank loan growth.
The bottom line is that there is no substitute for a deeply researched and well written piece. Buy side PMs often look at the most bullish and most bearish research pieces on a particular company or sector to understand the pros and cons of investing in a sector. A well written research report can be valuable even if ultimately the hypothesis is proven wrong.
Originally posted at Developing Insights Blog. The Developing Insights Blog (DI) is dedicated to emerging and frontier markets, which collectively will account for the majority of world output by 2020.
In Frontier and Emerging markets the cost of not understanding the big picture can be very painful. Countries with weak institutions, frequently coupled with very narrow export bases, can change quite rapidly on both the upside and the downside. This creates the potential for big losses in USD terms as well as numerous missed investment opportunities.
Graham and Dodd-style investing is predicated on the notion that value is recognized over the long term. As Benjamin Graham famously asserted, “in the short run, the market is a voting machine but in the long run, it is a weighing machine.” Implicit in that formulation are many developed market assumptions about political stability, exchange rate volatility, peaceful power transitions and a meaningful degree of economic policy consensus. Often, in a developing market context, those factors simply can’t be held constant. The ability to pursue a purist bottom-up strategy, devoid of political economy considerations and the macroeconomic fallout from poor policy, is severely constrained. If all risk is simply the variability of future outcomes then these additional “macro” considerations necessarily widen the band of potential returns for individual companies.
My last blog post focused on some tips on forecasting which were fairly qualitative . I thought that it would be great to follow up by discussing a few technical mistakes done during company valuation. While trying to forecast company performance and valuing companies, we often fall victim of what I would refer to as the ‘routine trap’. That happens when we apply certain methods without understanding the logic behind those assumptions even though situations and circumstances could be very different and require changes in methods. Thus we end up with inconsistent company valuation. Let me go a bit deeper into few common areas of mistakes. Be warned that this will be a fairly long post.