After introducing the inflation expectation, this paper uses the co-integration test and VAR model to analyze the price fluctuation of agricultural products and the paper analyzes the relationship between inflation and inflation expectation. The results show that there is no co-integration relationship between agricultural product price fluctuation, inflation expectation and inflation, but agricultural product price fluctuation is Granger reason of inflation expectation. There is bi-directional Granger causality between inflation expectation and inflation. In the short run, there is volatility between the three in the current or lag phase 1 to reach the maximum.
At present, the domestic scholars on the relationship between agricultural prices, inflation expectations and inflation are mostly concentrated in agricultural product price fluctuations and inflation causal analysis, the majority of the study that agricultural price fluctuations are contributing to inflation. The research is helpful for us to deepen the understanding of the relation between agricultural product price and inflation, but these researches still have the following shortcomings: Firstly, we do not take into account the effect of agricultural price change on inflation expectation, the expected impact of agricultural price fluctuations is the most direct, and this expectation has on the macroeconomic operation and the government’s macro-control has had a great impact, most scholars in the research process is ignored this important part ; Secondly, there is no agricultural price fluctuations, inflation expectations and inflation into a dynamic system for analysis, in-depth study of the interaction between the three [
The empirical research in this paper mainly analyzes the relationship between agricultural commodity price fluctuation, inflation expectation and inflation from the following two aspects. First, the long-run equilibrium angle using the method of co-integration analysis studies the interaction between them. The second is the short-term dynamic relationship, mainly using impulse response function and variance decomposition method for impact response and contribution analysis.
Vector Auto Regressive (VAR) is a model based on the statistical properties of data. The idea is to construct each model as a function of the hysteretic of all endogenous variables in the system. The model is often used in predicting interconnected time series systems. It is also commonly used to analyze the dynamic impact of stochastic disturbances on variable systems and to explain the impact of various economic shocks on the formation of economic variables [
Where
In this paper, the monthly price index data from January 2011 to November 2015 was used to analyze the long-term and short-term relationship between agricultural commodity price fluctuation, inflation expectation and inflation. The relevant variable description and source are as follows:
1) Price index of agricultural products (AP). This paper selects the retail price index of food retail price index as the proxy variable of the agricultural price index. The retail price index of food is a relative number that reflects the trend and fluctuation of the retail price of food market in a certain period. It directly affects the residents’ feelings about the price fluctuation, which affects their expectation.
2) The level of inflation expectations (IE). In this paper, consumer expectations index is chosen as the proxy variable of inflation expectation. It is the expectation of consumer’s change in future economic life. It is the general consumer’s income, savings, macro economy, consumption expenditure, employment situation, purchasing durable consumer goods and quality of life in the next year is expected and the next two years in the purchase of housing and renovation, the purchase of cars and the next six months the stock market is expected to change [
3) Inflation Index (IR). Using the Consumer Price Index (CPI) selected by most of the studies to represent the level of inflation.
The above data are derived from the National Bureau of Statistics website and the CCER China Economic and Financial Database, in order to eliminate seasonal patterns, so that changes in the sequence of time points to better reflect the underlying laws of the data, so that the statistics of different seasons. The data are comparable, this paper uses the X-11 method for seasonal adjustment [
When using co-integration test to carry out empirical analysis, it is required that all the time series of co-integration test must meet the conditions of horizontal instability and same order differential stationary. Therefore, we must first test the stability of the variables. In this paper, the ADF (Augmented Dickey-Fuller test) unit root test method to test the stability of the various time series variables, test results in
From
Since each variable sequence is a first-order differential stationary series, the Johansen-Juselius co-integration test can be used to determine whether there is a long-term co-integration relationship among the three variables.
(1) Determination of the lag order p
An important issue in the VAR model is the determination of the lag order. In
Inspection form | Test value | Critical value | Conclusion | Inspection form | Test value | Critical value | Conclusion |
---|---|---|---|---|---|---|---|
LnAP (0,0,0) | 0.4704 | −2.6054 | Non-stationary | ΔLnAP (0, 0, 0) | −6.7424 | −2.6062 | stationary |
LnIE (0, 0, 0) | −0.8439 | −2.6054 | Non-stationary | ΔLnIE (0, 0, 0) | −6.0323 | −2.6062 | stationary |
LnIR (0, 0, 3) | −2.4080 | −3.5550 | Non-stationary | ΔLnIR (0, 0, 2) | −2.0725 | −1.9469 | stationary |
Note: The test form brackets represent the constant term, the trend term and the lag order, lag order determined by the SIC and AIC criteria, Δ said first-order difference operator.
this paper, the Lagrange Criterion is used to evaluate the most reasonable VAR model to establish the lag time. It can be seen from
(2) Granger test (Granger Causality Tests)
In order to study the interaction between DLnAP, DLnIE and DLnIR, Granger causality test is carried out in this paper. The test results are shown in
It can be seen from
(3) Co-integration test
In order to test the existence of long-term stable equilibrium relationship between price fluctuation of agricultural products, inflation expectation and inflation level, the Johansen co-integration test based on regression coefficient is used to test the above variables. The results are shown in
The results in
Lag | LogL | LR | FPE | AIC | SC | HQ |
---|---|---|---|---|---|---|
0 | 539.7987 | NA | 0 | −20.2566 | −20.1450* | −20.2136* |
1 | 550.7889 | 20.3215* | 2.97e−13* | −20.3316* | −19.8856 | −20.1601 |
2 | 557.3173 | 11.3322 | 0 | −20.2384 | −19.4577 | −19.9382 |
Note: * indicates the minimum lag time given by the criterion.
Null hypothesis | F value | Probability value |
---|---|---|
DLnIE is not a DLnAP Granger cause | 1.7556 | 0.1908 |
DLnAR is not a DLnIE Granger cause | 4.6290 | 0.0359 |
DLnIR is not a DLnAP Granger cause | 2.7732 | 0.1016 |
DLnAP is not a DLnIR Granger cause | 2.5141 | 0.1186 |
DLnIR is not a DLnIE Granger cause | 8.9082 | 0.0043 |
DLnIE is not a DLnIR Granger cause | 3.1577 | 0.0812 |
Note: After the original logarithm sequence after the differential DLnAP = LnAPt − LnApt−1, this represents the price of agricultural products fluctuations.
Null hypothesis | Characteristic root | Trace statistics (p. value ) | |
---|---|---|---|
There are 0 cointegration relations | 0.1770 | 4.841466 (0.5622) | 18.44280 (0.6564) |
There are 1 cointegration relations | 0.1024 | 1.331289 (0.5611) | 3.531692 (0.6062) |
There are 2 cointegration relations | 0.0192 | 1.0880 (0.2969) | 1.0880 (0.2969) |
Note: The above test contains a constant but not a trend.
In order to further analyze the interaction of the three variables in short-term changes on each other, the following will use impulse response function and variance decomposition for analysis. Since there is no co-integration relationship among the three variables, this paper is based on the VAR (1) model, which is based on the first order difference of the logarithm sequence of three variables.
(1) Impulse response function analysis
Impulse response function (IRF) characterizes the effect of the change or impact of each endogenous variable on its own and all other endogenous variables and reflects the dynamic characteristics of the system. The specific shock response diagram is shown in
the expected standard deviation of a shock, the reaction of inflation is positive, in the first 2 to reach the maximum, and then began to slow down; and from their own impact, the reaction of inflation in the current maximum of 0.025, began to weaken, also in the first 7 to close to 0.
It can be seen that the response of the impulse response is consistent with the theoretical expectation. The response of agricultural price fluctuation to the impact from inflation expectation and inflation is positive, but the effect is relatively weak, and the reaction of inflation expectation to a shock from agricultural price is also positive, indicating that when the price of agricultural products up, the residents in the short term inflation expectations will follow the uplink, but the impact lasted only about three or so disappeared.
(2) Variance decomposition analysis
The variance decomposition describes the relative importance of the impact of each variable in the VAR model on the dynamic changes of the system variables. The main idea is to decompose the system’s prediction mean square error into its contribution rate of other shocks, so as to understand the relative importance of each variable impact on model endogenous variables. We then proceed to variance decomposition of the variables in the VAR model to further examine the factors that affect the price volatility of agricultural products, inflation expectations and inflation.
From
indicating that agricultural price fluctuations are mainly from their own impact.
It can be seen from
The variance decomposition shows that inflation expectation and inflation contribute less to the fluctuation of agricultural product price, while the contribution rate of agricultural product price fluctuation to inflation fluctuation is the largest, and the fluctuation of inflation expectation is explained by its own fluctuation.
This paper makes an empirical study on the relationship between the price fluctuation of agricultural products, inflation expectations and inflation in China between January 2011 and November 2015. The main conclusions are as follows:
(1) There is no long-term stable equilibrium relationship between agricultural product price fluctuation, inflation expectation and inflation, but there exists Granger causality between them, among which there is one-way causal relationship between agricultural product price fluctuation and inflation expectation. That is, the fluctuation of agricultural product price fluctuation can Granger cause inflation expectation change, but the change of inflation expectation can not be significant. Granger causes the fluctuation of agricultural product price, which is consistent with our expectation. It proves that the rise of agricultural price can cause inflation expectation; there is a two-way Granger causality between the expected change and the change of inflation, and there is no Granger causality between agricultural price fluctuation and inflation fluctuation. This conclusion also agrees with the research results of some scholars, which shows that the price increase of agricultural products only one form of inflation, rather than the cause of inflation, the impact of rising agricultural prices on inflation is passed through the impact of inflation expectations, because the residents of the demand for agricultural products is rigid, and its price increases will have a direct impact the residents of the inflation expectations. Although inflation expectations are not equal to the actual inflation, but inflation expectations will continue to accumulate, resulting in spiraling increases in commodity prices and inflation expectations, and ultimately evolved into significant inflation.
(2) Using the impulse response function and the variance decomposition to analyze the fluctuation conduction intensity and efficiency between the three, the results show that the inflation expectation and the fluctuation of inflation have a lag in the transmission of the agricultural price fluctuation, and their transmission effect with time to strengthen, but the effect is not obvious. The impact of agricultural price fluctuations and inflation shocks on inflationary expectations in the second period reached the maximum effect, while the impact of inflation on agricultural products from the impact of price fluctuations in the current period to reach the maximum effect, and this effect is positive. But since the first period after the rapid decline, in addition, inflation from the impact of inflation expectations in the current period there is a positive response, lagging behind a maximum.
In summary, although the long term, agricultural prices, inflation expectations and inflation does not exist between the stable equilibrium relationship, but the agricultural price fluctuations are inflation expectations of Granger reasons, and inflation expectations and inflation. There is also a two-way Granger causality, and the transfer between the three short-term fluctuations in the transfer effect is still more obvious, especially agricultural price fluctuations on the role of inflation in the short term more significant. In addition, although the impact of agricultural price fluctuations on inflation expectations is not very significant, but to some extent, or exacerbated the formation of inflation expectations.
Combined with the actual situation of China’s current stage and the conclusions of empirical analysis, this paper presents the following recommendations:
(1) The impact of agricultural price fluctuations on inflation in the short term is obvious. Therefore, it is necessary for the government to strengthen the regulation of the market, maintain the stability of agricultural prices and avoid a sharp increase in inflation, on the one hand by guiding the market to stimulate the enthusiasm of farmers to ensure the supply of agricultural products and enhance supply confidence, on the other hand to control the current agricultural products to be hype the market risk, to avoid causing a wider range of acts of price hikes; see, control of China’s inflation, it should not start from the inhibition of agricultural prices.
(2) The rise of agricultural prices on the formation of inflation expectations will have a positive impact, and inflation expectations will continue to accumulate and promote inflation, it is imperative to strengthen the management of inflation expectations, not only from the stability of agricultural prices to manage inflation expected, but also should strengthen the scientific guidance of public opinion, and guide people to rationally deal with price increases, to avoid blindly follow the wrong message, exacerbate the formation of inflation expectations.
Gou, G.H. (2017) The Study on the Relationship between Agricultural Product Price Fluctuation and Inflation Journal of Service Science and Management, 10, 166-176. https://doi.org/10.4236/jssm.2017.102015