Hancock Farmland Investor

Transcription

1 Hancock Farmland Investor Second Quarter 2016 Farmland and Timberland: Working Together in a Mixed Asset Portfolio A combination of farmland and timberland expands the potential to reduce variance in total return Farmland and Timberland assets have been used and tracked as components of institutional portfolios for over two decades, providing historically strong performance, low to moderate risk, and favorable diversification characteristics. In general, farmland and timberland have been managed separately and not in an integrated fashion. Yet, both farmland and timberland are income generating and land appreciation investment vehicles with biological growth components, offering comparable riskadjusted returns and inflation protection. Evaluating and structuring coordinated investments in these two natural resources has the potential of generating operational efficiencies and augmenting the risk-reducing diversification of a broader portfolio. This article provides a comparison of the risk-return profile of a combined farmland/timberland vehicle together with commercial real estate and other financial assets. Further, we compare performance results over the past twenty-five years for pure farmland and pure timberland to a pro-forma combined timberland/farmland vehicle. Historically, investments in farmland have provided annual total real returns (net of inflation) of 12 percent on average, while timberland total returns have averaged 10.6 percent, adjusted for inflation. Both farmland and timberland have historically provided a relatively high rate of return for their associated level of risk compared with other asset classes. To illustrate the potential benefits of a coordinated investment across these two natural resource classes, we constructed a Pro-Forma combined farmland/timberland investment vehicle based on historical return performance for assets in the United States. Our modeled farmland/timberland vehicle consists of 50 percent farmland and 50 percent timberland (rebalanced quarterly), and reflects the geographic, species mix and age-class distribution of the timber properties reporting into the National Council of Real Estate Investment Fiduciaries (NCREIF) Timberland Index, and similarly, the geographic distribution and the mix of row and permanent crops underlying NCREIF s Farmland Property Index. Chart 1 illustrates the risk-return profile for a combined farmland/timberland vehicle, compared to pure farmland, pure timberland, commercial real estate, and various financial assets during the past twenty five years ( ). The total return for the combined farmland/timberland vehicle is positioned Hancock Agricultural Investment Group Second Quarter 2016 Chart 1: Historical Risk vs. Return for U.S. Asset Classes ( ) 50/50 Farmland and Timberland Sources: NCREIF, HNRG Research, Macrobond between the individual historical returns for farmland and timberland, yet has a lower volatility than either of its components. The standard deviation of the combined farmland/ timberland vehicle dropped to 6.2 percent versus 9.1 percent for pure timberland and 6.7 percent for pure farmland. Viewing return performance on a risk-adjusted basis highlights the benefits of the combination of the two natural asset classes. Using the Sharpe Ratio as a measure of risk-adjusted performance, the combined farmland/timberland vehicle, when compared with traditional investment options (equities, fixed income, commercial real estate, and commodities), achieved the strongest risk-adjusted performance with a Sharpe Ratio of 1.36, followed by farmland at 1.35, and timberland at 0.83 (see Table 1, page 2). Investment returns for farmland and timberland exhibit a strong correlation of roughly 0.5, but are far from perfect substitutes. These two natural resources have performed differently under specific economic and policy conditions and are sensitive to different market drivers. To illustrate the differences in performance between farmland and timberland over a variety of periods, we have constructed a comparison over three time (Continued on page 2)

2 Farmland and Timberland: Working Together in a Mixed Asset Portfolio Continued from page 1) Table 1: Historical Risk vs. Return for U.S. Asset Classes ( ) Asset Class Sharpe Ratio Asset Class Sharpe Ratio Pro-Forma 50/ S&P Farmland 1.35 Int'l Equities 0.13 Timberland 0.83 Treasury Bills 0.00 Corporate Bonds 0.63 Commodities Commercial R.E *Risk Free Rate Used: Avg. 30-day T- Bill over Small Stocks period (2.76%) Sources: Macrobond, NCREIF, HNRG Research periods: the first time period, starts at the earliest point in time when synthetic returns were available for both farmland and timberland 1 ; covers the period from when NCRIEF reported performance returns became available for the asset classes up through the peak of the Global Financial Crisis (GFC); and finally , the period following the GFC. We have also included historical performance of NCREIF s Commercial Real Estate Property Index (NPI) for comparison as a point of reference, as it is the single largest category of institutional real-asset investment historically. Over extended periods of time, nominal total property-level returns for farmland, timberland, and commercial real estate have all stayed, on average, in the high single-digit to low double-digit range. However, performance for each of the three real assets has experienced significant variance from the average in particular time periods. In the first period ( ), timberland returned an average 14.9 percent, its strongest performance in all three periods, while farmland registered its lowest average returns. In the decade and a half preceding the GFC ( ), farmland and timberland both posted moderate returns of 11.3 percent and 12.2 percent respectively, Chart 2: Comparison of Real Asset Performance (Average Percent per Year) while commercial real estate dropped to an average of 7.2 percent. In the wake of the GFC, timberland returns dropped sharply to an average of 5.7 percent, reflecting a historic collapse in U.S. residential construction activity and an exceptionally lackluster housing recovery in the post-gfc period. However, farmland returns were exceptionally strong following the GFC, averaging 14.4 percent, and showing limited vulnerability to the global economic slowdown. Supporting robust returns for farmland investments in the period were a variety of factors, including: U.S. government mandated use of ethanol in car fuel coupled with trade restrictions limiting U.S. imports of ethanol; historically high commodity prices; and strong Chinese imports of agricultural commodities. Mixed together in equal proportions, a combined farmland and timberland portfolio showed extremely consistent return performance across all three distinctly different periods. Offsetting each other s weak performance periods, the combined farmland and timberland portfolio had an average total return over the three periods of 11.1 percent, with a spread between the highest and lowest average return over the three periods of 170 bps. This compared favorably with U.S. commercial real estate which had an average total return over the three periods of 10.4 percent, with a spread between the highest and lowest average return over the three periods of 520 bps. Table 2: Historical Average Nominal Return (Percent per Year) U.S. Timberland U.S. Farmland Pro-Forma 50/50 Farmland/ Timberland U.S. Commercial Real Estate 14.9% 8.3% 11.6% 11.8% 12.2% 11.3% 11.7% 7.2% 5.7% 14.4% 10.1% 12.3% Sources: NCREIF, HNRG Research Sources: NCREIF, HNRG Research The return information from the chart above is summarized in Table 2 and illustrates the complementary nature of farmland and timberland in a coordinated investment vehicle. A combined farmland and timberland portfolio brings strong diversification benefits when combined with equities and fixed income assets. Chart 3, on page 6, presents a fifteen year rolling correlation of annual returns of a 50/50 farmland/ timberland portfolio to equity and bond returns. Over the entire 25 year period, a combined farmland and timberland portfolio averaged a 0.09 correlation to the S&P 500, and a negative 0.27 correlation to long-term corporate bonds (Morningstar Ibbotson U.S. long-term corporate bond index). (Continued on page 6) Hancock Agricultural Investment Group Second Quarter

3 Quarterly Corn Production Estimates, Major Producers Sources: USDA WASDE. Note: 2015 data are estimates and 2016 data are projections, updates are made on a corn marketing year basis. Figure 1. U.S. corn production is estimated at MMT for the 2015/16 marketing year, a 4% decline from 2014/15 due to lower yields and harvested acres. 2016/17 corn production is projected to increase to MMT due to a 7% increase in harvested acres and good weather conditions during the growing season. Brazil s 2015/16 corn crop decreased by 21% from 2014/15 to 68.5 MMT due to poor yields, but 2016/17 production is expected to rebound to 82.5 MMT on higher acreage and better yields. Argentina s 2015/16 projected corn production of 28 MMT represents a 2% decrease from 2014/15. Yet, 2016/17 corn production is projected to increase to 36.5 MMT as farmers took advantage of the elimination of corn export taxes and planted more acreage to corn. China s corn production in 2015/16 is estimated at MMT and is projected to fall to 216 MMT in 2016/17 due to the removal of corn price supports. Quarterly Soybean Production Estimates, Major Producers Figure 2. Global soybean production is forecast to increase 5.5% in 2016/17 due to increased production in Brazil, the U.S., and India. Brazil s soybean production is estimated to decrease slightly to 96.5 MMT in 2015/16 due to a reduction in planted area. In 2016/17, planted area in Brazil is projected to increase and boost the country s soybean production to 101 MMT. Argentina s soybean production is estimated to decrease by 7% to 56.8 MMT in 2015/16 and stay flat in 2016/17 as export tax policies favor corn production. U.S. soybean production in 2015/16 is estimated to increase slightly from 2014/15 production to MMT. U.S. soybean production is projected to continue increasing (7%) in 2016/17 to MMT as a result of increased harvested acreage and strong yields. Sources: USDA WASDE. Note: 2015 data are estimates and 2016 data are projections updates are made on a soybean marketing year basis. Quarterly Exchange Rates Between USD and Agricultural Currencies (Indexed to 1 at 2006: Q1) Figure 3. The USD appreciated against most foreign currencies in Q as a result of strong U.S. economic data and expectations of an interest rate increase. However, the currencies of Russia and Brazil appreciated against the dollar from the previous quarter, by 5% and 10% respectively. When comparing the strength of the USD to competitive currencies over the last year: the Brazilian real depreciated 3% against the USD; the Russian ruble depreciated 13% against the USD; the Canadian dollar depreciated 4% against the USD; the Argentine peso depreciated 39% against the USD; the Australian dollar depreciated 3% against the USD. Sources: Macrobond, HNRG Research Hancock Agricultural Investment Group Second Quarter

4 Four Quarter Moving Average Corn Exports, Major Producers Figure 4. Brazil s four quarter moving average corn export volume increased by 74% since last year, boosted by the weak Brazilian real but negatively impacted by the drought which affected the second crop, leaving second quarter volumes flat over first quarter as supply tightened. Argentina s four quarter moving average volume of corn exports decreased by 8% since last quarter and decreased by 4% since last year. U.S. four quarter moving average corn export volume in Q increased by 5% quarter-over-quarter largely due to reductions in corn supplies in Brazil, but decreased by 3% year-over-year. Sources: Argentinian Ministry of Agroindustry, Aliceweb, USDA, HNRG Research Quarterly export data is a four quarter moving average Four Quarter Moving Average Soybean Exports, Major Producers Figure 5. The U.S. four quarter moving average soybean export volume remained flat from last year and from last quarter. Brazil s four quarter moving average soybean export volume increased by 4% from last quarter and has increased by 31% year-over year. Argentina s four quarter moving average soybean export volume fell by 34% since last quarter and decreased by 12% year-over-year. Argentina s export sales have been negatively affected by lower production and by farmer s holding back soybean stocks in hopes of a further reduction in soybean export taxes. Sources: Argentinian Ministry of Agroindustry, Aliceweb, USDA, HNRG Research Quarterly export data is a four quarter moving average Quarterly Corn and Soybean China Imports Figure 6. Chinese imports of corn and soybeans totaled 25 MMT in Q2 2016, a 17% increase over Q China s soybean imports typically peak in quarters two, three, and four due to product availability. Chinese soybean imports are projected to reach record levels in 2016 as domestic production will not meet the increasing demand from domestic livestock and poultry, along with demand for vegetable oil consumption. Chinese corn imports in Q increased by 61% against Q Corn imports have surged as a result of spoiled grain inventories and inexpensive corn prices in the international market. Source: China General Administration of Customs Hancock Agricultural Investment Group Second Quarter

5 Quarterly U.S. Row Crop Prices ($USD per bu) Figure 7. U.S. corn and soybean prices increased in Q corn prices increasing by 3% since last quarter and by 1% since last year as a result of increased exports and concerns over drought conditions in Brazil. Soybean prices increased by 12% since last quarter and by 1% since last year as a result of strong export demand. Year-over-year, Q wheat prices declined by 20%, and by 5% since last quarter as a result of the strong dollar, increased global production, and competition, particularly from the EU and the Black Sea region. Source: USDA NASS, Prices reported as received by buyer Annual U.S. Average Grower Tree Nut Prices ($USD per lb.) Figure 8. The U.S. is a major producer of tree nuts, accounting for approximately 82% of the world s almonds, 28% of the world s walnuts and 23% of the world s pistachios in U.S. prices for almonds, walnuts and pistachios fell in the 2015 marketing year (August-July for almonds and July-June for walnuts and pistachios). Almond prices fell 29% year-over -year, walnut prices fell 52% and pistachio prices declined by 31%. Driving the decline in almond prices has been the strong U.S. dollar and a continued caution from buyers, a reaction to the high prices in Walnut prices were affected by weak exports, foreign competition, a large crop, and carry-over stocks in The 2015 U.S. pistachio crop declined dramatically from 2014, yet prices still fell as a result of weak export demand. Sources: USDA NASS, Prices reported as received by buyer, Nichols Farms, BLS, HNRG Research tree nut prices are estimates, official prices to be reported by the USDA in July 2016 NCREIF Row Crops Total Return (percent per year) and USDA Row Crops Cash Receipts (y/o/y percent change) Figure 9. Row crop cash receipts are projected to decline for the fourth consecutive year in 2016 as declining corn and wheat prices offset a slight increase in soybean prices. Row crop cash receipts are forecast to fall by 3.3% in 2016 reaching $157 billion. The largest declines in row crop cash receipts are expected to come from feed crops, followed by food grains, vegetables, and melons. NCREIF s year-to-date Q row crop total return of 2.4% were slightly higher than year-todate Q row crop total return of 2.1% but was still the second lowest year-to-date second quarter return for NCREIF row crops on record. Row crop total returns are expected to respond to downward pressure from the projected decline in USDA row crop cash receipts in Sources: USDA ERS, NCREIF Farmland Index 2015 USDA row crop cash receipts are an estimate, 2016 USDA row crop receipts are a forecast Hancock Agricultural Investment Group Second Quarter

6 Farmland and Timberland: Working Together in a Mixed Asset Portfolio (Continued from page 2) Operational Efficiencies and Expanded Opportunity Set A coordinated approach to incorporating farmland and timberland into an institutional portfolio has the potential to expand the opportunity set of properties targeted for acquisition and provide greater flexibility in building a set of investments to meet individual investment objectives. Broadening the search for properties across farmland and timberland allows more flexibility in assembling a portfolio of properties and avoiding overheated markets in a particular market segment or geography. The flow of large scale, high quality farmland and timberland properties to the market is neither smooth nor continuous, and having a broader mandate across both farm and timberland would enhance an investors ability to act more opportunistically. Farmland and timberland encompass a wide array of crop (row and permanent) and forest types producing for a multitude of end-use markets. The depth and diversity of the investment options that both farmland and timberland offer are a key factor in their ability to bring diversification and risk reduction to an investment portfolio. The above review of the historical performance and investment characteristics of farmland and timberland suggest that a combined management approach when incorporating these two asset classes in a broad institutional portfolio has the potential to reduce the volatility of returns and facilitate the property acquisition process. Chart 3: Combined Farmland/Timberland Portfolio 15-Year Rolling Correlation Sources: NCREIF, HNRG Research Note: 1 Timberland returns prior to 1987 are the Hancock Timberland Index, a synthetic nominal total return series based on historical timber prices and assumed capitalization rates. Early farmland returns, prior to 1991 are based on methodology within the 2009 study by Francis and Ibbotson; Contrasting Real Estate with Comparable Investments. HNRG Research Team Court Washburn Managing Director and Chief Investment Officer cwashburn@hnrg.com Keith Balter Director of Economic Research kbalter@hnrg.com Bill Devens Senior Agricultural Economist bdevens@hnrg.com Mary Ellen Aronow Senior Forest Economist maronow@hnrg.com Elizabeth Shestakova Economic Research Analyst eshestakova@hnrg.com Keith Goplerud Economic Research Analyst kgoplerud@hnrg.com NOTES: Figure 1. Corn production is charted on a calendar year basis and updated on a marketing year basis. The corn marketing year is from September to August for the U.S., from May to April for South Africa, and from October to September for China. The corn marketing year is from March to February in Argentina and Brazil. Corn Production data and forecasts are updated on a monthly basis by the USDA World Agricultural Supply and Demand Estimates Report (WASDE). Figure 2. Soybean production is charted on a calendar year basis and updated on a marketing year basis. The soybean marketing year is from September to August for the U.S., from February to January for Brazil and for April to March for Argentina. Soybean Production data and forecasts are updated on a monthly basis by the USDA World Agricultural Supply and Demand Estimates Report (WASDE). Figure 3. Exchanges rates are updated on a daily basis by Macrobond Financial AB. Figure 4. Argentina s agricultural exports are published on a monthly basis by the Argentinian Ministry of Agroindustry. Brazil export data is published on a monthly basis by Aliceweb. U.S. exports are published on a monthly basis by the U.S. Census Bureau. Export data is reported on a 4 quarter moving average to adjust for seasonality. Figure 5. Argentina s agricultural exports are published on a monthly basis by the Argentinian Ministry of Agroindustry. Brazil export data is published on a monthly basis by Aliceweb. U.S. exports are published on a monthly basis by the U.S. Census Bureau. Export data is reported on a 4 quarter moving average to adjust for seasonality. Figure 6. China s soybean and corn imports are published on a monthly basis by the China General Administration of Customs. Figure 7. Row Crop Prices are published on a monthly basis by the USDA National Agricultural Statistics Service (USDA NASS). Figure 8. Permanent Crop Prices are published on a annual basis by the USDA National Agricultural Statistics Service (USDA NASS). Walnuts price estimates for the current year are calculated by using the percent annual changes for the crop year in the Producer Price Index for walnuts, published by the Bureau of Labor Statistics on a monthly basis. Annual estimates for pistachios are calculated by using the percent annual changes for the crop year in the prices published in the Market Reports from Nichols Farms. Almond price estimates for the current year are calculated by using the percent annual changes for the crop year in the prices from HNRG sources. Figure 9. USDA Cash Crop Receipts data is published three times a year in February, August and November by the USDA s Department of Agriculture Economic Research Service. The U.S. level calendar-year forecast is first published in February. The August release converts the previous year s forecast to estimates and the November release updates the current year forecast. NCREIF Farmland Total Return data is published on a quarterly basis. NCREIF quarterly total row crops returns are aggregated to form the total return for the year. The total return as seen on the bar chart may not equal the annual total return as reported by NCREIF, because the NCREIF annual return is calculated by multiplying instead of adding quarterly returns together. References to expected investment performance in this newsletter are based on historical information and are based on managements projections. Potential for profit as well as for loss exists. Hancock Agricultural Investment Group is a division of Hancock Natural Resource Group, Inc., a registered investment advisor and wholly owned subsidiary of Manulife Financial Corporation. Hancock Natural Resource Group, Inc. Hancock Agricultural Investment Group Second Quarter