CC BY
106. Удалова Т.Н. Экономическая эффективность лизинга на воздушном транспорте. Автореф. дис. на соиск. учен. степ. канд. экон. наук /08.00.05/. М., 1999.
7. Федяшова Е.А. Лизинг на воздушном транспорте в условиях возрастающих международных требований к эксплуатации воздушных судов гражданской авиации. Дис. на соиск. учен. степ. канд. экон. наук /08.00.05/. М., 2007, 135 с.
8. Хачатрян Г.А. Лизинг самолетов как хозяйственная деятельность // Вестник Национального университета архитектуры и строительства Армении. Ереван, 2014, 4 (42).
Стр. 121-130. (на армянском)
9. Човушян Э.О. Лизинг в России. М., 2006.
10. Чубаров Н.Н. Лизинг авиационной техники как инструмент повышения конкурентоспособности воздушного транспорта Российской Федерации. Дис. на соиск. учен. степ. канд. экон. наук /08.00.05/. Ростов-на-Дону, 2006, 191 с.
11. A320 Family. The Market Leader.- Airbus Leading aircraft manufacturer, June 2015. http://www.airbus.com/ aircraftfamilies/passengeraircraft/a320family/. 14.07.15.
HUMAN CAPITAL AS AN ASSET CLASS IN FINANCIAL PLANNING
Valentyn Khokhlov,
CFA
PhD, International marketing manager Global Spirits, Kiev, Ukraine
ABSTRACT
Recent advancements in financial planning require taking human capital into consideration when choosing an optimal portfolio for individual investors. For most individuals it's likely to be their single largest asset. This article refines the approach to defining human capital and relates in to the unemploy-ment-adjusted future labor income. Yet, human capital cannot be considered in isolation from other asset classes, our research reveals a significant correlation between increases in wages and lagged stock market returns. For people who have fixed amount of labor (e.g. full-time employees) the return on human capital is irrelevant for asset allocation, but the correlation of labor income and stock market affects choosing the optimal portfolio for individuals. However, it's the standard deviation of labor income that plays the key role in the allocation of financial assets.
Keywords: human capital, labor income, asset allocation, financial planning.
Introduction
The development of individual financial planning and portfolio management techniques during last decades has resulted in much more sophisticated approaches to asset allocation than ad-hoc equity/bond splits. However, introduction of mathematical modeling requires more accurate accounting for all the significant variables for decision making in asset allocation. As Bodie, Merton and Samuelson state, "ignoring human capital constitutes an 'omitted variable' problem" and "accounting for human capital is crucial to explaining investment, labor, and consumption behavior of rational economic agents" [1, p. 446].
Following the seminal paper by Bodie, Merton and Samuelson [1], it has become essential to consider the life-cycle model that takes human capital and labor income into consideration when making portfolio choice decisions. Koo [2] studies the asset allocation problem for agents who receive labor income, are liquidity constrained and have uninsured labor risks. He uses CRRA utility function in a continuous-time setting. A structural model of optional life-cycle consumption behavior was subsequently developed by Gourinchas [3]. Chen [4] includes life insurance into the optimal portfolio selection process while still using CRRA utility function. Their his research the authors assume that labor income is correlated with the stock market returns, while not especially stating the risk-return characteristics and the basis for determining correlation.
Subsequent developments refine and enhance the mathematical models accounting for human capital and labor income. Ehrlich [5] investigates the impact of education and
the opportunity cost of asset management on optimal portfolio choice. Ren [6] studies the predictive power of human capital and derived variable on the risky assets prices, and also they introduce a slightly different approach to defining human capital based on finite-state Markov chains. Finally, there are a distinct subset of research dedicated to the impact of skills and occupations on human capital and labor income, for example, Yamaguchi [7] and Silos [8] address this issue.
The objective of this paper is to investigate human capital characteristics as an asset class, particularly the properties of labor income and its correlation with the stock market. The paper is organized as follows. We start with the definition and estimation of human capital based on U.S. Bureau of Labor data. In the next section we examine the dependencies between labor income and the stock market returns. Finally, we analyze the risk and return characteristics of human capital as an asset class. The paper concludes with a brief summary of our findings.
The Definition of Human Capital
The financial economics value of human capital, following [1, p. 428] and [4, p. 105], is defined as the present value of future economic benefits that can be reasonably attributed to the person's skills and abilities. We consider the sum of (1) wages and salaries and (2) self-employed income as the proxy the monetary value of these benefits. This sum is called labor income, as opposed to non-labor income that includes dividends, interest, rental income, social security, public assistance, retirement benefits, etc.
Mathematically, the financial economics value of human capital (HC) is represented as
HC, = £
E[h ]
(1 + r )-
HC
(1)
the probability of unemployment:
E [h ] = E [ht\Ae ] P ( ) + E [ht\Au ] P ( Au )
(2)
where ^^ is the amount of HC at the beginning to year t, h. is the amount of labor income in year i (random variable), r is the discount rate, T is the retirement age (last year before retirement).
Chen [4, p. 105] calculates HC using the real after-tax wages or salary, and the relevant discount rate in that case is the real risk-free rate adjusted for illiquidity. While it's certainly a plausible assumption when HC considered per se, it's not entirely aligned with treatment of other sources of capital in asset allocation. Most academics and practitioners use pretax nominal returns on equity and fixed income securities, especially for generic asset allocation purposes. We believe that it's easier to use the nominal before-tax cash flows and discount rates whenever possible, so we suggest using the nominal before-tax labor income in the numerator and the relevant nominal before-tax discount rate in the denominator.
Another difference in our approach is the treatment of uncertainty and illiquidity of labor income. Chen uses the uncertain h, and adds the relevant risk premium to the discount rate. However, it's hard to estimate this add-on to the risk premium. Hence we suggest converting the uncertain hi to a certain one and using the relevant risk-free discount rate using
where
Ae, Au
mean the state of being employed and
E [h \ A ]
unemployed in year i respectively, L 1 e J is the expected value of labor income for an employed person (i.e.
E [ h\A ]
wage, salary, or self-employed income), L 1 "J is the expected value of labor income for an unemployed person (which is zero, because our definition of labor income does not
P (A )
include unemployment benefits), u is the probability
a person in unemployed in year (which is the unemployment
rate), and
P ( Ae ) = 1 - P ( Au ).
E [ MAe ]
In order to calculate L 1 1 e J we first take the sum of lines "Wages and salaries" and "Self-employment income" from [9] for the consumption units (CU) at available ranges and divide it by the number of earners in CU (see Table 1), getting the average employed person's labor income (LI) at several reference ages (shown with diamonds on Figure 1). Then we use quadratic interpolation to generate labor income at other ages (shown with dashed line on Figure 1).
Table 1.
Average U.S. labor income per earner in 2013
Reference age, years 21.6 29.8 39.7 49.7 59.2 68.8
Average earners in CU 1.3 1.5 1.6 1.6 1.3 0.5
Average LI per earner $19,277 $37,805 $47,104 $46,166 $48,190 $34,858
Source: Consumer Expenditure Survey (2013)
However, since our main data source for wages is the SSA we calibrate labor income so that the average labor income series [10], the average wage in 2014 should be S44,569. Hence from age 21 to age 65 constitute S44,569.
$60,000 T---------------------------------------------------------------------------------------
y = -38.205X2 -i- 3760x - 42704 R2 = 0.9567
$50,000 -$40,000 -
<5
<U
<U
a.
m $30,000 -«
M re
ai >
<
$20,000 £ $10,000 --------
$0
/
P (A )
We take the average unemployment rate u from
the Labor Force Statistics series "(Seas) Unemployment Rate"
20 30 40 50 60
Age
Figure 1. Average U.S. labor income during the working life-cycle
P (Au ) = 5.9%
1951-2014,
nominal cash flows,
E[hi]
70
As we have opted to use the
needs to be adjusted for inflation.
The average inflation rate, based on CPI series in 1872-2014, is 2.25%. Thus, the expected labor income of an average person aged i is calculated using the equation (2) adjusted accordingly:
E[h ] = Wtk(l - P(A, ))(1 +1)'-20 (3)
where W« is the approximated wage for an employed person
aged i as per Figure 1 (W = -39.205'2 + 376° -42704
)> k = 1.0639 is calibration constant to convert to 2014 $1,200,000
wages, P(A ) = 5.9%
is the average unemployment rate, 1 = 2.25% is the average inflation rate. We assume here that the person starts working at 21 year.
HC can be calculated using equation (1) with the nominal risk-free discount rate set as the average rate of return on 10-year U.S. Treasury Bonds r = 4.5% . Based on our assumption, HC at the age of 21 is $1,135,818, and it declines almost linearly during the working life-cycle until the age of 65 (see Figure 2).
re
<j
E
re
E
Age
Figure 2. Human capital over the working life-cycle
The Correlation between Labor Income and the Stock Market Return
Labor income (LI) by our definition is the sum of wages, salaries and self-employed income. Davis [11, p. 41] assumes that LI is uncorrelated with the stock market return, and that has a major implication on asset allocation. However, for us this assumption seems counter-intuitive, since salaries and, especially, performance bonuses for many employees depend on the overall state of the economy. Moreover, poor stock market performance is frequently accompanied by layoffs, which increases unemployment. Combined effect of having
E ] and P (A )
lower
in (2) negatively affects
E [ h ]
the effective expected labor income L ' J. Other academics acknowledge the existence of this interrelation, for example Chen [4, p. 107] introduces the correlation variable but does not provide a specific value.
In order to determine the correlation we use the SSA
series [10], which covers the wages and salaries, because self-employed income time series are unavailable. However, the former constitutes 93-94% of LI. So we regress either average
E [ h | A ]
actual wages (a proxy for L ' e * ) or average effective
E [ h ]
wages (a proxy for L 'J) against (1) the same-period stock market return, (2) the one-year lagged stock market return, (3) the one-year and the two-year lagged stock market returns. The S&P 500 price-only index return is used as a proxy for the stock market return (we have also investigated using the S&P 500 total return index as a proxy and found no material difference in the outcomes).
As we can see from Panel A on Figure 3, the same-period S&P 500 returns reveal no correlation with the actual average wage increases, which supports Davis conclusion [11, p. 41]. However, on Panel B we plot the wages increases against previous years S&P 500 returns, and there is a strong positive relationship.
Panel A 7% -,
re
ftt
41 00 re
(V <30
re
41 >
<
R2 = 0.0134
-60%
Panel B 7% -,
40%
-2% J -2% J Same-year S&P 500 return One-year lagged S&P 500 return
Figure 3. The relationship between actual averages wages increase and S&P 500 returns (1993-2014)
The regression analysis for the three models is provided in Table 2. The same-period S&P 500 returns have little explanatory power over the wages increases, and we can conclude their impact is immaterial. However, both one-year and two-year lagged S&P 500 returns are significant even at 1% level, so the wages depend on the past stock market performance, and the explanatory power of the model increases as we add the second previous year into consideration.
Table 2.
Labor income regression on S&P 500 (1993-2014)
Panel A
Model E [ht \Ae ] = axrt + b E [ht \ Ae ] = airt-l + b E [ht \ Ae ] = airt-1 + a2rt-2 + b
Variable Estimate (t- statistics) al b a1 b a1 a2 b
0.0112 0.0318 0.0666 0.0281 0.0639 0.0324 0.0260
(0.5208) (7.1797) (4.5139) (9.0973) (4.7636) (2.3957) (8.6066)
R2 Correlation 0.0134 0.1157 n/a 0.5175 0.7193 n/a 0.6436 0.7193 0.4090 n/a
Panel B
Model E [ht \Ae ] = airt + b E [ht \ Ae ] = airt-1 + b E [ht \ Ae ] = airt-1 + a2rt-2 + b
Variable Estimate (t-statistics) a1 b a1 b a1 a2 b
0.0203 0.0318 0.1041 0.0253 0.0999 0.0470 0.0218
(0.6645) (5.0486) (5.1209) (5.9631) (5.4426) (2.5432) (5.2633)
R2 Correlation 0.0216 0.1470 n/a 0.5799 0.7615 n/a 0.6956 0.7615 0.4063 n/a
Source: author's calculations
As we switch from the actual to the effective average wage increases, i.e. take into account the level of unemployment, the situation does not change materially — the same-year S&P 500 returns on the scatter plot (Panel A on Figure 4) are uncorrelated, but one-year lagged S&P 500 returns exhibit
significant correlation with wages increase (Panel B on Figure 4). However, as the regression analysis shows (see Panel B in Table 2), the explanatory power of the model increases as we switch to the effective (unemployment-adjusted) labor income.
Panel A
Panel B
8%
6%
> <
4%
-60%
-4%
-6%
20%
40%
Same-year S&P 500 return Orie-year lagged S&P 500 return
Figure 4. The relationship between effective averages wages increase and S&P 500 returns (1993-2014)
When analyzing the regression models with high explanatory power, we can clearly see the intercept is significantly different from zero and all estimates lie in the range from 2.2% to 2.8%, which is rather close to the long-term inflation rate of 2.25% we
used in (3). If we added the same-period inflation rate into the regression (see Table 3), the intercept becomes insignificant for the effective average wage model, but not for the actual average wage model.
Table 3.
Labor income regression on S&P 500 and inflation (1993-2014)
Model E [ht | ] = axrt-i + a2It + b E [ht ] = a!rt-1 + a2It + b
Variable Estimate (t-statistics) a1 a2 b a1 a2 b
0.0631 0.5016 0.0164 0.0992 0.6903 0.0093
4.4924 1.8382 2.3470 5.1301 1.8378 0.9630
R2 Correlation 0.5937 0.7193 0.3718 n/a 0.6462 0.7615 0.3592 n/a
Source: author's calculations
Human Capital in the Asset Allocation The asset allocation decision is about choosing the optional weights of different asset classes in a portfolio. We consider three classes: equities (risky assets), government bonds (risk-free assets), and HC. In the absence of the latter, the problem is a traditional problem of portfolio selection for
individual investors, and it's mostly solved with mean-variance
80%
70%
'5
£ 60% ■M Y5
S 50%
d <
1 40% £ a
30%
■6
I 20% +-
re
v
= 10% <
optimization (MVO) by adjusting the risk tolerance based on individual circumstances, including age. An example of such an asset allocation is shown on Figure 5, it starts with 75% allocated to equities at 25 years and ends with 20% allocated to equities at 65 years, with the balance allocated to bonds. It can be considered a "typical" allocation for a person.
- 1.4
- 1.2
- i
u
- 0.8 I
41 ß
- 0.6 a
E
- 0.4
- 0.2
-Allocation to equities - Risk tolerance
25 30 35 40 45 50 55 60 65
Age
Figure 5. Asset allocation ignoring human capital
In our definition of HC we assume the person is either to HC is fixed and the labor income to be independent of this working full-time or is unemployed. That means the allocation allocation. The alternative would be to assume a non-fixed
allocation, which implies floating working hours and parttime employment not only during the current year but also in future. However, since it's not the case for most persons and the complexity of non-fixed allocation is great, we do not address this case here.
To model a fixed allocation to HC we set the minimal and maximal weights for this asset class to the current weight, which equals to the ratio of HC calculated as per (1) to the total assets. The return of HC mathematically is calculated as
rHC =
h
HC
(4)
where ht is the amount of labor income in year t (random variable), HCt is the amount of HC at the beginning of year t. For solving the MVO problem, we use the expected return
HC
on HC,
Vt
HC
E [ ht ] HCt
. Unlike other
asset classes,
V HC
is not constant and depends on t. However, 100%
our research shows that the MVO solution actually doesn't depend on this input. It can be explained solely by the fact the weight of HC is fixed, so having a different expected return doesn't change the allocation to his asset class. We can put any
u HC
constant t into the model and get the same results.
While the expected return on HC is irrelevant for asset allocation, the risk plays a significant role. First of all, we have already addressed the correlation between labor income and the stock market return, which is approximately 0.76. The correlation with government bonds is not significant. The only remaining parameter needed for MVO is the standard deviation of HC, which is defined as the standard deviation of labor income (SDLI). We don't have any reliable research on this matter, however, so we investigate the optimal allocation of financial assets (because the allocation to HC is fixed) based on different values of SDLI. Using the same varying risk tolerance as on Figure 5, we solved a series of MVO problems for different values of SDLI (see Figure 6).
90%
~ 80% --
70% —
60%
50% —
40%
30%
20% —
10%
0%
V ■ V \ \ . *\
\ \ v
-SDLI 4.2% \ * ^ A __________________________• _____\ ________________________________
• • • • SDLI 6% \ v
— . -SDLI 9% \ \ \\ \ . t\
----SDLI 12%
- • ■ SDLI 15% ___
--SDLI 20%
/
/ -,-,-<-T-T-T-T-,
25
30
35
40
45 Age
50
55
60
65
Figure 5. Asset allocation considering human capital
For the most part of feasible range (4.2% to about 20%) the optimal allocation of financial assets is 100% equities until 42-52 years, and then it quickly declines to 20% at 65 years. It can be explained by the fact that HC behaves like a bondlike assets with low SDLI, so high allocation to HC crowds out any further allocation to bonds, but as HC declines bonds appear in the optimal portfolio. However, as SDLI approached the risk of equities HC becomes more equity-like and crowds out allocation to equities, so the optimal portfolio at ages 2540 becomes 100% bonds, and only then some equity allocation starts to appear and as the amount of HC declines. In that case the allocation of equities tops out at 50-55 years at the level of 30%, and then declines to 20% at the retirement age.
Conclusions
Human capital plays an important role in the asset allocation decisions of individual investors. For a typical person it represents a very significant part of her portfolio, especially in the early years. Ignoring this dominant asset class can easily mislead allocation between risky and risk-free financial assets.
Our research reveals a significant correlation (approximately 0.76) between labor income and one-year lagged stock market
returns. So HC cannot be considered as uncorrelated with equities. Asset allocation decisions should take it into account. However, even more critical input is the standard deviation of labor income. It plays a crucial role in asset allocation, because for its low levels human capital may be considered a bond-like asset even when it's highly correlated with equities, but high levels of this standard deviation make human capital an equitylike asset. On the other hand, the expected return on human capital is irrelevant for asset allocation decisions when the weight of human capital is fixed.
References:
1. Bodie, Zvi, Robert C. Merton and William F. Samuelson. 1992. "Labor supply flexibility and portfolio choice in a life cycle model." Journal of Economic Dynamics and Control. 16 (3): 427-449.
2. Koo, Hyeng Keun. 1998. "Consumption and Portfolio Selection with Labor Income: A Continuous Time Approach." Mathematical Finance. 8 (1): 49-65.
3. Gourinchas, Pierre-Olivier and Jonathan A. Parker. 2002. "Consumption over the life cycle." Econometrica. 70 (1): 47-89.
4. Chen, Peng, Roger G. Ibbotson, Moshe A. Milevsky and portfolios." Review of Economic Dynamics. 18 (3): 635-652. Kevin X. Zhu. 2006. "Human Capital, Asset Allocation, and 9. Consumer Expenditure Survey, U.S. Bureau of Labor Life Insurance." Financial Analysts Journal. 62 (1): 97-109. Statistics, September, 2014. URL: http://www.bls.gov/cex/2013/
5. Ehrlich, Isaac, William A. Hamlen Jr. and Yong Yin. 2008. combined/age.xlsx (Accessed on 19 Jan 2016).
"Asset Management, Human Capital, and the Market for Risky 10. National Average Wage Index, U.S. Social Security
Assets." Journal of Human Capital. 2 (3): 217-261. Administration, 2015. URL: https://www.socialsecurity.gov/
6. Ren, Yu, Yufei Yuan, Yang Zhang. 2014. "Human capital, OACT/COLA/AWI.html#Series (Accessed on 19 Jan 2016). household capital and asset returns." Journal of Banking & 11. Davis, Steven J. and Paul Willen. 2000. "Occupation-Finance. 42: 11-22. Level Income Shocks and Asset Returns: Their Covariance and
7. Yamaguchi, Shintaro. 2012. "Tasks and Heterogeneous Implications for Portfolio Choice." NBER Working Paper No. Human Capital." Journal of Labor Economics. 30 (1): 1-53. 7905. URL: http://www.nber.org/papers/w7905 (Accessed on
8. Silos, Pedro and Eric Smith. 2015. "Human capital 19 Jan 2016).
ЗНАЧЕНИЕ ИНВЕСТИЦИОННОЙ ДЕЯТЕЛЬНОСТИ НА СОВРЕМЕННОМ ЭТАПЕ
РАЗВИТИЯ РОССИЙСКИХ ПРЕДПРИЯТИЙ
Чараева Марина Викторовна
доктор экономических наук, профессор кафедры «Финансовый менеджмент» Ростовского государственного экономического университета (РИНХ)
Дирацуян Михаил Хевонтович Магистрант магистерской программы «Корпоративные финансы» (направление «Финансы и кредит») Ростовского государственного экономического университета (РИНХ)
VALUE OF INVESTMENT ACTIVITY AT THE PRESENT STAGE OF DEVELOPMENT OF THE RUSSIAN ENTERPRISES
Charaeva Marina Viktorovna Doctor of Economics, professor Financial Management chairs Rostov state economic university (RINH)
Diratsuyan Mikhail Hevontovich Undergraduate of the master program "Corporate finance" (Finance and Credit direction) Rostov state economic university (RINH)
АННОТАЦИЯ
Цель - исследовать особенности инвестиционной деятельности российских предприятий в современных экономических условиях, исходя из чего определить значение инвестирования для российской экономики. В процессе работы применялись научные методы: анализ и синтез, индукция и дедукция, детализация, группировка, сравнение. Результат - обозначены перспективы инвестирования на российских предприятиях. Особо следует обратить внимание на то, что усложняется процедура экономической оценки инвестиционных проектов в связи нестабильностью макроэкономической ситуации. Выводы: постоянный мониторинг инвестиционного состояния российской экономики позволит повысить эффективность принятия стратегических решений по определению приоритетных направлений инвестирования.
ABSTRACT
The purpose - to investigate features of investment activity of the Russian enterprises in modern economic conditions proceeding from what to define value of investment for the Russian economy. In the course of work scientific methods were applied: analysis and synthesis, induction and deduction, specification, group, comparison. Result - investment prospects at the Russian enterprises are designated. Especially it is necessary to pay attention that procedure of an economic assessment of investment projects in communication by instability of a macroeconomic situation becomes complicated. Conclusions: continuous monitoring of an investment condition of the Russian economy will allow to increase efficiency of adoption of strategic decisions on definition of the priority directions of investment.
Ключевые слова: инвестиционная деятельность, российские предприятия, экономическая оценка, инвестиционные проекты.
Keywords: investment activity, Russian enterprises, economic assessment, investment projects.
Российский рынок - один из самых привлекательных проблема актуальна прежде всего тем, что на инвестициях для иностранных инвесторов, однако он также и один из в России можно нажить огромное состояние, но в то же самых непредсказуемых. Инвесторы крайне чутко реаги- время боязнь потерять вложенные средства останавлива-руют на экономическую ситуацию в регионе и не стре- ет инвесторов, хотя они всеми силами пытаются не упу-мятся рисковать своим капиталом. Россия относится к стить свою долю российского рынка и, в то же время, не странам повышенного риска по многим причинам. Эта потерять свои деньги. При этом иностранные инвесторы
