Economics · Model
The RBC Model with a Stock Market
What does the canonical RBC model say about the stock market? In this version of the model, the firm owns the capital stock and households own the firm: they hold its shares, which pay the dividends the firm does not invest. Without adjustment costs, the allocation is exactly that of the canonical model, so the model delivers a stock price and a stream of dividends at no extra cost, and they can be compared with the S&P 500 and with the dividends paid in the US economy. The comparison is a classic test of the model, and a demanding one (Rouwenhorst, 1995; Jermann, 1998).
The Model
Households
The household’s problem
A representative household consumes \(c_t\) and divides one unit of time between leisure, \(l_t\), and hours of work, \(h_t\), so that \(l_t + h_t = 1\). Its utility is logarithmic in consumption, so the coefficient of relative risk aversion is one, which is consistent with balanced growth (King, Plosser and Rebelo, 1988), and \(\theta\) is the weight of leisure. It saves by holding shares of the representative firm, \(S_t\), bought at the ex-dividend price \(p_t\) and paying the dividend \(d_t\):
With \(\lambda_t\) the multiplier on the budget constraint, the first-order conditions give \(\lambda_t = 1/c_t\), labour supply and the Euler equation for shares:
where \((p_{t+1} + d_{t+1})/p_t\) is the gross return on the firm’s stock.
Firms
Production, investment and the firm’s problem
A representative firm owns the capital stock and hires hours to produce with a Cobb–Douglas technology:
Technology has two parts. Total factor productivity, \(A_t\), fluctuates around its long-run level with persistence \(\rho\), and labour-augmenting technical progress, \(X_t\), grows at the constant rate \(g_X\) each quarter and gives the economy a balanced growth path:
The firm pays wages and invests out of its revenue and distributes the rest as dividends; there are no adjustment costs, so a unit of output buys a unit of capital:
The firm maximises its value to its shareholders, the expected stream of dividends discounted with their stochastic discount factor, \(\beta^t \lambda_t/\lambda_0\), by choosing hours, investment and capital:
With \(q_t\) the multiplier on the accumulation of capital (Tobin’s \(q\)), the first-order conditions are
Installed capital is worth exactly what it costs, \(q_t = 1\), and the firm invests until the discounted return on a unit of capital equals one.
General Equilibrium
Definition of competitive equilibrium
Given the initial capital stock, \(K_0\), the household’s initial portfolio, \(S_{-1}\), the initial level of technology, \(A_0\), and its stochastic process, a competitive equilibrium consists of state-contingent sequences of household choices (consumption, labour supply and share holdings), firm choices (labour demand, investment, capital and dividends) and prices (the wage and the share price), in which every variable dated \(t\) depends only on the history of shocks up to \(t\), such that:
- Household optimisation. Given prices, households maximise lifetime expected utility subject to their sequence of budget constraints.
- Firm optimisation. Given prices and the shareholders’ stochastic discount factor, the firm maximises the expected discounted value of its dividends subject to its flow-of-funds constraint and the accumulation of capital.
- Market clearing. In every period and for every history of shocks, the labour market clears, the household holds the whole stock of the firm, \(S_t = 1\), and the goods market clears: combining the two budget constraints, \(c_t + I_t = Y_t\).
Labour supply, the Euler equation for shares and the resource constraint are then the three equations of the canonical model, in which households own the capital and rent it to firms: who owns the capital does not matter when markets are complete and there are no adjustment costs. The allocation, consumption, hours, investment and capital, is the same; what this model adds is the price of the firm and its dividends.
Stock Price and Dividends
The value of the firm equals its capital stock
With one share outstanding, the Euler equation for shares prices the firm as the discounted value of next period’s price and dividend, \(p_t = \beta\, \mathbb{E}_t [(\lambda_{t+1}/\lambda_t)(p_{t+1} + d_{t+1})]\). Because the technology has constant returns to scale and there are no adjustment costs, the ex-dividend value of the firm is the replacement cost of its capital (Hayashi, 1982), and dividends are capital income less investment:
To check the first, substitute \(p_{t+1} + d_{t+1} = K_{t+2} + \alpha Y_{t+1} - I_{t+1} = \alpha Y_{t+1} + (1-\delta) K_{t+1}\) into the pricing equation: the firm’s first-order condition for capital then gives \(p_t = K_{t+1}\). The return on the stock is the return on capital, \((p_{t+1} + d_{t+1})/p_t = 1 + \alpha Y_{t+1}/K_{t+1} - \delta\).
Two predictions follow. The stock price moves only with the capital stock, which changes slowly, so it should be smooth and very persistent. And dividends fall when investment rises: in a boom, investment rises by more than capital income, so dividends are countercyclical.
Methodology
The model is calibrated as the canonical RBC model and driven by the same technology shocks, measured in the US data, quarter by quarter, from the same starting values. Its consumption, hours, investment and capital are those of the canonical model; the stock price and dividends follow from them (see “Stock price and dividends in the stationary economy” in Definitions, Methods and Solutions). The settings below choose the technology series and the shocks used everywhere on the page.
These settings apply to every chart and table below. Linear trend: the shocks of King and Rebelo (1999), with the model’s series HP filtered afterwards. HP-filtered: the shortcut, with no filtering of the model’s series. One comparison at a time.
Calibration
The preferences, technology and trend growth are calibrated as for the canonical RBC model: King and Rebelo (1999, Table 2), with logarithmic utility, a discount factor implied by the average real return on the S&P 500 (the model’s stock return is the return on capital) and a weight of leisure implied by hours of 20 percent of available time (see the Calibration page), and the trend, persistence and size of the technology shocks estimated from the technology series selected above.
Technology shocks
The technology shocks, and the starting values of technology and capital in 1947 Q1, are those of the canonical model; see Technology shocks on the canonical RBC model page.
Steady State
The steady state
Dividends, the value of the firm and the return on stocks
On the balanced growth path, variables divided by \(X_t\), written with a tilde, are constant. The accumulation of capital gives \(\tilde{I} = (g_X + \delta)\, \tilde{K}\), so dividends and the value of the firm are fixed shares of output:
The gross return on stocks equals that on capital, \(R = (1+g_X)/\beta\), and the price–dividend ratio is \(\tilde{p}/\tilde{d}\), a quarter of the ratio measured with annual dividends, as in the data.
Steady state and great ratios
The steady state of the calibrated model, with trend growth, and the average of the price–dividend and dividend–price ratios of the S&P 500 since 1950.
Model Simulations vs the Data
The model is fed with the measured technology shocks, quarter by quarter, and its output, stock price and dividends are compared with the HP cyclical components of the US data. The stock price is the real S&P 500 index: a price per share, which is not divided by population. The model’s dividend, \(d_t = \alpha Y_t - I_t\), is the payout of the whole capital stock, so its counterpart in the data is the aggregate net dividends of US corporations (national accounts), shown by default, in total or per person like output. The dividends per share of the S&P 500 are shown for comparison: being per share, they are the same in both views. The switches also apply to the statistics below.
Business-Cycle Statistics: Model and Data
The statistics of King and Rebelo (1999, Table 1) for output, the stock price and dividends, in the US economy and in the model, over the same quarters.
Correlation with the data
How much of the US history the model reproduces: the volatility of each model series relative to the data, and the correlation between the two, quarter by quarter.
The model misses on both counts. Its stock price barely moves and is far too persistent, because the value of the firm is its capital stock; in the data, stock prices are several times more volatile than output (Shiller, 1981). Its dividends are more volatile than in the data and strongly countercyclical, while US dividends are smoother and mildly procyclical, because firms smooth them (Lintner, 1956). Explaining the stock market requires adjustment costs, which make the price of capital move, and preferences or shocks that make risk premia vary over time (Jermann, 1998; Boldrin, Christiano and Fisher, 2001; Campbell and Cochrane, 1999).
Impulse Responses to One Positive Technology Shock
The quantities respond as in the canonical model. The stock price follows capital: it rises gradually, peaks with the capital stock after several years and returns slowly to its path. Dividends fall sharply on impact, because the firm invests most of its extra income, then recover gradually and, after about five years, rise above their path, as investment returns to normal while the larger capital stock keeps capital income high.
Definitions, Methods and Solutions
Equilibrium conditions
Combining labour supply and labour demand, the Euler equation for shares with the firm’s condition for capital, and the two budget constraints with \(S_t = 1\), the allocation solves the three equations of the canonical model in consumption, hours and capital:
and the stock price and dividends follow: \(p_t = K_{t+1}\) and \(d_t = \alpha Y_t - I_t\).
Stock price and dividends in the stationary economy
Dividing by \(X_t\), \(\tilde{p}_t = p_t/X_t = (1+g_X)\, \tilde{K}_{t+1}\), and the accumulation of capital, \((1+g_X)\, \tilde{K}_{t+1} = (1-\delta)\, \tilde{K}_t + \tilde{I}_t\), gives the stock price in terms of this quarter’s capital and investment. Log-linearised around the steady state, with hats for percent deviations,
Both are linear in the canonical model’s output, investment and capital, so they hold for the simulated series, before or after HP filtering (the filter is linear), and for the impulse responses. The coefficients, with the technology series selected above:
The canonical model is solved as on its page (Blanchard and Kahn, 1980). Investment is about three times as volatile as output in the model, so with these coefficients a 1 percent rise in output above trend comes with a fall in dividends of several percent.
Data
- Stock price: the S&P Composite index (the S&P 500 from 1957) from Robert J. Shiller’s data, averaged over each quarter and deflated by the GDP price index (GDPDEF), as on the Stock Market page. It is a price per share, not divided by population in either view.
- Dividends, total: twelve-month dividends per share of the S&P 500 (Shiller), deflated by the GDP price index.
- Dividends, per capita: net dividends paid by US corporations in the national accounts (DIVIDEND, U.S. Bureau of Economic Analysis), deflated by the GDP price index and divided by the population (B230RC0Q173SBEA). The model’s dividend is the total payout of the firm per person, so its counterpart is aggregate dividends per person, not dividends per share.
- Output: real GDP (GDPC1), in total or per person, the latter as on the canonical RBC model page.
All data are HP filtered (λ = 1600) as \(100 \times \ln x_t\) on their full history and shown from 1950 Q1.
References
- Blanchard, O. J. and C. M. Kahn (1980). “The Solution of Linear Difference Models under Rational Expectations.” Econometrica 48(5): 1305–1311.
- Boldrin, M., L. J. Christiano and J. D. M. Fisher (2001). “Habit Persistence, Asset Returns, and the Business Cycle.” American Economic Review 91(1): 149–166.
- Campbell, J. Y. and J. H. Cochrane (1999). “By Force of Habit: A Consumption-Based Explanation of Aggregate Stock Market Behavior.” Journal of Political Economy 107(2): 205–251.
- Hayashi, F. (1982). “Tobin’s Marginal q and Average q: A Neoclassical Interpretation.” Econometrica 50(1): 213–224.
- Jermann, U. J. (1998). “Asset Pricing in Production Economies.” Journal of Monetary Economics 41(2): 257–275.
- King, R. G., C. I. Plosser and S. T. Rebelo (1988). “Production, Growth and Business Cycles: I. The Basic Neoclassical Model.” Journal of Monetary Economics 21(2–3): 195–232.
- King, R. G. and S. T. Rebelo (1999). “Resuscitating Real Business Cycles.” In J. B. Taylor and M. Woodford (eds.), Handbook of Macroeconomics, Vol. 1B, Chapter 14: 927–1007. Amsterdam: Elsevier.
- Lintner, J. (1956). “Distribution of Incomes of Corporations Among Dividends, Retained Earnings, and Taxes.” American Economic Review 46(2): 97–113.
- Rouwenhorst, K. G. (1995). “Asset Pricing Implications of Equilibrium Business Cycle Models.” In T. F. Cooley (ed.), Frontiers of Business Cycle Research, Chapter 10: 294–330. Princeton University Press.
- Shiller, R. J. (1981). “Do Stock Prices Move Too Much to Be Justified by Subsequent Changes in Dividends?” American Economic Review 71(3): 421–436.
Page created on 16/11/2026.
© Arthur Galichère.
- Errors: Every effort has been made to ensure the model and derivations are accurate; any remaining errors are mine. If you believe you have found a mistake, or have a comment or suggestion, please let me know.
- Reuse and citation: Stock-market data are Robert J. Shiller’s (shillerdata.com); total factor productivity is from John Fernald at the Federal Reserve Bank of San Francisco; the other data are from the U.S. Bureau of Economic Analysis via FRED. The model simulations, charts and text are the author’s. When reusing these figures, please cite the original data providers and this website: Arthur Galichère, The RBC Model with a Stock Market.