Abstract. The aim of this paper is to examine the existence and persistence across various portfolios of stocks of the January effect, a market anomaly suggesting that security prices exhibit distortions in the first month of every year compared to the rest of the year. The existing empirical studies on the January effect in contemporary financial markets, particularly in recent years, are somewhat limited. To provide a comprehensive overview, this paper utilizes data encompassing all American stocks listed on NASDAQ, AMEX, and NYSE from 1980s to 2023. The selection of the American stock market as the sample is based on its complexity and substantial trading activity. This investigation delves into the anomaly by assessing its impact on stock portfolios categorized by size, book-to-market ratio, and dividend yield. Additionally, the paper scrutinizes cross-sectional variations within the portfolios to uncover potential patterns explaining this anomaly, by looking at the corresponding quintile distribution. The data undergoes regression analysis using dummy Ordinary Least Squares. The results of the analysis reveal that the anomaly affects only a subset of stocks that share common characteristics: small capitalization, high book-to-market value and the ones that did not yield investors any dividends. The results are in line with previous findings in literature, particularly those of Ross (1983) and Reinganum (1983), suggesting evidence in favour of the window dressing hypothesis and the tax-loss selling hypothesis.

Keywords: efficiency, stock market, yield, econometric testing.