COVID-19疫情控管程度會改變主要資產對抗通貨膨漲的能力嗎？動態縱橫門檻Fisher模型之應用

Can the control degree of the COVID-19 epidemic change the ability of major assets to hedge against inflation? An application of dynamic panel threshold Fisher model.

陳奕勳(YI-XUAN CHEN)

指導教授 ： 吳博欽 ； 劉曉燕

中原大學/商學院/國際經營與貿易學系/碩士(2022年)

https://doi.org/10.6840/cycu202201552

摘要

本研究建構一個以嚴格指數為門檻變數的動態縱橫門檻Fisher 模型 (dynamic panel threshold Fisher regression model, DPTFR)，評估主要資產對抗通貨膨脹時所產生的長、短期門檻效果。實證上，採用最具代表性的加密貨幣－比特幣、傳統的保值資產－黃金，以及傳統的風險性資產－股票為對象進行估計。實證期間則為COVID-19疫情爆發與持續期間，亦即2020年1月至2021年12月的週資料，共計有306筆觀察值。
實證結果摘錄如下：
(一)在任何疫情的嚴重性、警戒級數與嚴格指數水準下，比特幣、黃金與股票等資產雖無法對抗通貨膨脹，卻存在門檻效果，且門檻變數 (嚴格指數, SI) 的門檻值為 20.370。
(二)當SI小於20.370時，警戒級數大多集中在一級，少數在二級，而資產名目報酬的遞延效果為0.2343，且資產短期對抗通貨膨脹的能力為-3.4722，而長期對抗通貨膨脹的能力則為-4.5347。此外，物價上漲而使資產名目報酬受到傷害的程度較傳統的Fisher模型之估計結果更為嚴重。
(三)當SI大於20.370時，警戒級數大多集中在二級與三級，資產名目報酬的遞延效果為-0.0138，且資產短期對抗通貨膨脹的能力為-0.2754，而長期對抗通貨膨脹的能力則為-0.2717。此外，因物價上漲而使名目報酬受到傷害的程度則較傳統的Fisher模型之估計結果略為下降。
投資決策與政策建議如下：
(一)面對歷史未曾出現過的重大疫情衝擊，資產投資者無法透過重要資產投資以對沖通貨膨脹。因此，趨避風險投資者對資產投資應選擇觀望的態度。
(二)在疫情爆發的初期 (SI大多低於20.370)，投資者應避免大量拋售所持有的該等資產，以免資產報酬受通貨膨脹的侵蝕更大。在疫情爆發的中、後期 (SI大多高於20.370)，資產對抗通貨膨脹的能力仍為負值，而投資者卻可在此階段降低所持有的該等資產，以減輕資產報酬受通貨膨脹的侵蝕。
(三)政府必須積極地採取相關的政策，一方面降低通貨膨脹的持續惡化，大幅侵蝕資產的實質報酬，另一方面穩定資本 (股市) 與金融市場 (比特幣與黃金市場)，避免資產價格大幅下挫，進而導致資金與資本外移，影響經濟發展。

關鍵字

費雪假說 ； 嚴格指數 ； 動態縱橫門檻Fisher 模型 ； 依時與狀態變動 ； 非線性縱橫單根檢定 ； 通貨膨脹率

並列摘要

This study constructs a dynamic panel threshold Fisher regression (DPTFR) model with a stringency index as the threshold variable to evaluate the long-term and short-term threshold effects of major assets against inflation. Empirically, the most representative assets are used for estimation, including cryptocurrency (bitcoin), traditional value-preserving asset (gold), and traditional risk asset (stocks). The empirical period includes the outbreak and duration periods of the COVID-19 pandemic, that is, the weekly data from January 1, 2020 to December 31, 2021, a total of 306 observations.
The empirical results are summarized as follows:
(1) Although bitcoin, gold, and stocks cannot hedge against inflation, under any level of the severity, alert grade, and stringency index of the pandemic, there have threshold effects, and the threshold value of the threshold variable (strict index, SI) is 20.370.
(2) When SI is less than 20.370, the corresponding alert grades are mostly concentrated in the grade I and a few in the grade II, and the deferred effect of the nominal return of the assets is 0.2343. The short-term effect for the assets against inflation is -3.4722, and the long-term effect is -4.5347. In addition, inflation hurts nominal returns on assets more severely than the result in the traditional Fisher model.
(3) When SI is greater than 20.370, the corresponding alert grades are mostly concentrated in the grade I and grade III, and the deferred effect of the nominal return of the assets is -0.0138. The short-term effect for the assets against inflation is -0.2754, and the long-term effect is -0.2717. Besides, inflation hurts nominal returns on assets more slightly than the result in the traditional Fisher model.
Relevant investment strategies and policy suggestions are as follows:
(1) In the face of a major epidemic that has never occurred in history, asset investors cannot hedge against inflation by investing in important assets. Thus, risk-averse investors should take a wait-and-see attitude towards asset investment.
(2) In the outbreak stage of the COVID-19 pandemic (SI is mostly below 20.370), investors should avoid selling these assets in large quantities, so as to avoid asset returns being eroded by inflation. In the middle and late stages of the pandemic (SI is mostly above 20.370), the ability of assets against inflation is still negative, but investors can reduce their holdings of these assets to reduce the erosion of asset returns from inflation.
(3) The government must actively adopt relevant policies, on the one hand, to reduce the continuous deterioration of inflation, which greatly erodes the real return of the assets, on the other hand, to stabilize the capital market (stock market) and financial market (bitcoin and gold market), and avoid asset prices sharply decline, which in turn led to the outflow of funds and capital and influence economic development.

並列關鍵字

Fisher hypothesis ； stringency ； dynamic panel threshold Fisher regression model ； time and state vary ； nonlinear asymmetric heterogeneous panel unit root test ； inflation

參考文獻

Alagidede, P., and Panagiotidis, T. (2010). Can common stocks provide a hedge against inflation? Evidence from African countries. Review of Financial Economics, 19(3), 91-100.

Anari, A., and Kolari, J. (2002). House prices and inflation. Review of Estate Economics, 30(1), 67-84.

Armantire, O., Kosar, G., Pomerantz, R., Skandalis, D., Smith, K., and Topa, G. (2020). Inflation expectations in times of COVID-19. Federal Reserve Bank of New York, Liberty Street Economics. Available at https://libertystreeteconomics.newyorkfed.org/2020/05/inflation-expectations-in-times-of-covid-19.html.

Arnold, S., and Auer, B.R. (2015). What do scientists know about inflation hedging? North American Journal of Economics and Finance, 34, 187-214.

Aye, G.C., Edoja, P.E, and Charfeddine, L. (2017). Effect of economic growth on CO2 emission in developing countries: Evidence from a dynamic panel threshold model. Cogent Economics & Finance, 5(1), 1-22.