What is the right benchmark for your strategies?
The most appropriate benchmark for comparison with regard to our Large Cap Active Equity product is either the S&P 500 Index or the Russell 1000 Index. The S&P 500 is the customary benchmark used by our clients; however, we also have clients that benchmark us against the Russell 1000. Either is acceptable to us.
The most appropriate benchmark for comparison with regard to our Mid Cap Active Equity product is either the Russell Midcap Index or the S&P MidCap Index. We have clients that benchmark against either the S&P or the Russell Midcap indices. Either is acceptable to us.
The most appropriate benchmark for comparison with regard to our Small-Mid Cap Active Equity product is the Russell 2500 Index or the Russell 2000 Index.
What is the Alpha objective of your strategies?
New Amsterdam Partners’ performance goal for our strategies is to generate alpha of 4% net of fees per year and a tracking error in the 3 to 8% range.
What time period does your expected return model cover?
The model gives expected return for the next twelve months.
What are the Tracking Error, Information Ratio and Beta of the Large Cap and Mid Cap portfolios?
New Amsterdam Partners’ portfolios are actively managed. As such, our Large Cap and Mid Cap portfolios show a tracking error in the 5.0 to 8.0% range and an information ratio in the 0.3 to 0.7 range (over rolling five and ten-year periods). Our historical beta has remained relatively low: in the 0.8 to 1.0 range.
What is the smallest market cap you will buy in the portfolio(s)?
The smallest market cap we would buy in any portfolio is $200 million for our Small-Mid Cap Active Equity strategy.
What is the largest market cap you will buy in the portfolio(s)?
Our Small-Mid Cap and Mid Cap portfolios are cap-constrained; we will only consider buying companies with market caps less than $3.5 billion at the time of purchase for the Small-Mid Cap strategy and less than $9 billion at the time of purchase for the Mid Cap strategy. For Large Cap portfolios, we buy names with an initial market cap greater than $5 billion.
Does the model consider different economic sectors?
The model is run in two ways: once unconstrained and once with sector bands. The output of both model runs groups the companies by their sector. The unconstrained model run is free to select the top 100 names regardless of sectors. This tells us which sectors the model likes or dislikes, based on how many stocks it is picking up from each sector (i.e., sector weight). The second model run is constrained such that the sector weights (number of names in a sector) are at least 50%, and no more than 200%, of the weights in the index. These sector weight constraints are consistent with our portfolio building process.
How do you approach sector diversification?
Our portfolios have approximately 40 to 45 stocks diversified across 10 economic sectors (we define our sectors using GICS classifications). By investing across 10 sectors we seek to reduce the risk of overexposure to any single segment of the economy. For major sectors, we are never more than two times or less than one half the benchmark weighting. Weights for smaller sectors may fall outside this band at times but will not tilt the portfolio due to the relative size of our major sector weights.
What sector(s) does each analyst cover?
Michelle Clayman, CFA, Managing Partner & Chief Investment Officer: – Chairs Investment Meetings – Determines the investment philosophy – Investment research; industry generalist and health care sector
Nathaniel Paull, CFA, Partner & Senior Portfolio Manager: Director of research – Portfolio structuring and trading/compliance supervision – Investment research in information technology and telecommunication services sectors
Hong Hoang, CFA, Principal & Senior Equity Analyst: Generalist, financials and information technology sectors, Quantitative research and backtesting
John Hingher, CFA, Principal & Senior Equity Analyst: Consumer sectors
Daniel Buslik, Junior Quantitative Analyst: Quantitative research and backtesting
How many securities does each analyst cover ?
Our Investment Group covers about 150 securities: our Buy Universe list (a result of our proprietary expected return model), including our current holdings, as well as certain securities on a “watch list” made up of stocks that have appeared on that list in the past.
Why do your portfolios hold 40-45 stocks?
Modern Portfolio Theory tells us that a fully diversified portfolio can be constructed with as few as 30 stocks. On the other hand, our practical experience puts the number more realistically at 40 to 45 stocks. Our process allows us to focus on the highest expected return stocks – to own more than the effective minimum to properly diversify would reflect “name creep,” and reduce the overall return delivered to our clients.
Do you have a back end model to control risk?
We monitor and control risk at each stage of our investment process and define risk not simply as a specific quantitative measure but also as methodology risk. We believe there is inherent risk in longstanding processes that are not back tested and carefully evaluated for enhancements.
Quantitative Research Risk Control: Our backtesting process is part of New Amsterdam Partners’ investment technology diagnostics. The process tests the validity of our investment model and indicates possible areas for improvement of both the model and our investment philosophy.
Backtests are run for a variety of purposes. We monitor correlations between our model-generated forecast expected returns and actual stock price performance to quantitatively evaluate the model’s predictive ability. We also analyze each attribute of our expected return model to reaffirm that all attributes contribute significantly to the model’s discriminatory power. The test confirms that the combination of components is more valuable than any single component used alone. Additionally, we evaluate the performance of New Amsterdam Partners’ actual stock picks versus the performance of model picks.
Fundamental Research Risk Control: When presenting a stock for purchase, each analyst must also complete the “Fundamental Checklist,” a detailed breakdown of time-tested issues our most senior analysts have found useful when evaluating a company. Some of these issues include: revenue recognition policy, debt load and maturity schedules, death spiral covenants, option accounting, pension liabilities, litigation concerns, effective tax rates, inventory bulges, off-balance sheet financing and corporate governance.
Analysts use FactSet Research Management Solutions (RMS) to structure, manage, and track fundamental analysis using proprietary formats. The RMS, through FactSet Internal Research Notes (IRN), allows for data that resides in FactSet and data that the analyst enters either through the IRN templates, which have been built, or by attaching documents to the electronic IRN, to be aggregated in a data downloaded worksheet. This system replaced an internal legacy system but incorporated many of the essential items. This allows us to see a full picture of the investment story of a company over time. Significant fundamentals are stored under the categories of, potential for positive and negative surprise, and risk factors. Quantitative statistics such as expected return and growth estimates are also stored for each company. The reports generated by our research platform capture company information, current market prices versus price targets, risk factors and themes within an industry.
Portfolio Risk Control: During portfolio construction, we look at risk control in three areas: stock-specific risk, stock diversification, and sector diversification. We examine each of our risk exposures versus those of our benchmark to make sure that our portfolios do not contain any unintended bets. Additionally, one of the things we focus on during the fundamental research stage of our investment process is whether or not purchase candidates have similar risk exposures to current holdings; by identifying these characteristics, we can avoid unintended bets in our portfolio.
Do you use an optimizer?
New Amsterdam Partners does not use an optimizer to structure or explicitly manage portfolios as we utilize a bottom-up stock selection process. However, we are concerned with both sector weightings and stock weightings within our portfolios, and place constraints around both. Additionally, while we do not utilize an optimizer to explicitly manage our portfolios, we do use optimization software to evaluate risk exposures. This allows us to understand and evaluate our risk exposures so that we do not have unintended bets within our portfolios.
What is your sell discipline?
Portfolio holdings are continually monitored for possible sale in a number of ways: First, we establish price targets for all holdings. Once reached, if there is no further expected upside, we sell. Secondly, underlying fundamentals are continuously evaluated, which can lead to an analyst revising his/her price target. Thirdly, our expected return model, identifies those holdings that have suffered a drop in expected return, due to deteriorating fundamentals. Conversely, it also highlights new securities that may offer
At what market cap does a mid cap stock exit the portfolio?
New Amsterdam Partners defines mid cap stocks as companies with market capitalizations between $1 billion and $9 billion. This cap restriction is placed on securities at the time of purchase. However, if the value of a stock appreciates so that its market cap exceeds $9 billion, we will not sell the position based on valuation alone, but rather on the determination of our Investment Group. As a stock appreciates, we systematically reduce its portfolio position if its value reaches 5.0% of the portfolio’s market value.
Does the team decide which stocks to buy/sell or the CIO?
All buy and sell decisions are made via a consensus between the Chief Investment Officer and Senior Portfolio Manager (with the CIO having final say). They rely on individual sector analysts to assist them in making these decisions.
The two portfolio managers talk daily with each other and interact with other members of the team as needed on developments that may impact the portfolio. The portfolio managers also meet regularly following each model production run.
How often is the model run?
Our proprietary expected return model is run twice a month. We may run the model at any other time as we feel necessary.
How do your fundamental analysts forecast a company’s cash flow?
Our cash flow forecast is a distillation of all that we know about a given company, its industry, and the current and long-term macroeconomic outlook. To understand a company and its financial productivity requires a number of steps. First, we examine all publicly available data, including annual and quarterly reports, paying particular attention to how the company has performed in times of both economic expansion and recession. We also examine the balance sheet relative to the cash that will be required to finance both the anticipated growth of the firm as well as the economic depreciation of the firm’s assets. Next, we look at management’s track record, including their ability to grow the business both organically and through the selective use of acquisitions. Finally, we analyze the industry itself, trying to discern whether or not it is an inherently attractive (shareholder-value creating) industry or one with less favorable long-run return characteristics.
Once we have a good understanding of these factors, we prepare to forecast the company’s annual cash flows. Our forecasts typically cover a whole business cycle, 7 to 10 years, and take into account what we know about how revenues and margins will be impacted by both a slowing and accelerating economy. We are conservative in our estimates and do not assume perpetual straight-line revenue growth; most businesses/industries demonstrate some form of cyclicality and we try to capture this in our forecasts.
What pick up in expected return do you look for when swapping out of one stock and into another?
If we decide to swap one stock for another in the same or similar sector, we would only consider candidates that offer a 4% pick-up in expected return after transaction costs.
What is your typical holding period for a stock / turnover?
Turnover in our portfolios is relatively low – in the 40 to 100% range over a rolling 12-month period – resulting in a typical holding period of three to four years. This illustrates our conviction in what we buy and aligns the long-term outlook of our portfolios with that of our clients’ investment horizons. A byproduct of this approach is a benefit to our tax-sensitive clients.
How are portfolio characteristics (p/e, p/b, growth, roe) calculated?
We calculate various characteristics for our composites, portfolios, and relevant indices as a portfolio risk control measure. The characteristics are in the areas of valuation, dividend yield, expected growth, total return, profitability, leverage, past performance, systematic risk, and market capitalization. The calculation of the characteristics is done using accounting and pricing data from Factset Data Systems. This whole process is automated using the Portfolio Analytics module in FactSet Data Systems. Portfolio characteristics are calculated as of market closing on each month-end.
Do you use Street estimates for forecast growth, return on equity, etc.?
We use Street estimates as a starting point. We then make a series of algorithmic adjustments to these estimates as they are inputted into our expected return model. These adjustments take into account an overestimation bias on the part of the Street as well as anomalous accounting. An important part of our process is to identify where the consensus earnings estimate is and how our view and insights differ – that is one way we add significant value. Over time, our backtesting shows our estimates to be more accurate than those of the Wall Street consensus.
How much cash is held in the portfolio?
Our portfolios are fully invested; cash is held only as a residual to our process or to accommodate client distribution needs and is typically at 3% or less, though it never exceeds 5%.