IPOs發行家數及每個月IPOs平均初始報酬都出現顯著的自我相關，這是所謂IPOs發行熱季的異象。許多公司當觀察到最近市場上的IPOs，有高初始報酬後，也會跟隨著上市。本文研究發現，申請時市場上任何IPOs的初始報酬，與隨後申請的IPOs其未來上市或上櫃後的初始報酬無關。當申請IPOs發行前，市場報酬爲負時，則發行公司會將市場資訊完全反應在定價上，此時大部分的初始報酬，可由IPOs從申請到正式發行期間的資訊來解釋。這段時間中，如果市場報酬增加或投資人有超額需求時，IPOs的初始報酬也隨著增加。此外計畫發行IPOs的公司，會認爲市場對於最近的IPOs有好的評價，而選擇趕快申請上市上櫃，但卻無法預測到折價的程度。這就是爲什麽存在高IPOs初始報酬後，IPOs發行量也增加的原因。

Ibbotson and Ritter (1995) indicate that there are three anomalies in IPO markets: IPO underpricing, IPO long-run underperformance, and hot-issue market. Contrary to most of the previous studies which focus on IPO underpricing or IPO long-run performance, this paper examines the existence of hot-issue phenomenon in Taiwan IPO markets. ”Hot-Issue anomaly” means that cycle exists in the both the IPO volume and IPO underpricing. Consequently, issuers tend to issue new equity to the public when faced with high average initial returns of IPOs. Ritter (1984) argues that IPOs in hot markets experience higher initial returns. Moreover, young IPOs experience more underpricing in hot markets. A possible explanation for hot-issue phenomenon is the positive feedback hypothesis: market investors positively react to IPO underpricing. When investors earn initial returns from IPOs, they are more likely to subscribe to future IPOs leading to the fact that issuers tend to issue when previous IPOs are more underpriced. Hot-issue is considered as an anomaly because underpricing or initial return is referred as an indirect cost of issuance. Issuers should try to reduce the extent of underpricing to raise IPO proceeds. From the point of view of maximizing proceeds, previous studies fail to explain the hot-issue anomaly in IPO markets. Ibbotson, Sindelar and Ritter (1994) show that IPO hot-issue phenomenon exists not only in U.S. market but also in Germany, South Korea, and U.K. markets. To make up this gap, this paper first tests if there exists hot-issue anomaly in Taiwan and then examines why hot-issue anomaly exists to provide an explanation to the anomaly. We find that IPO initial return leads IPO issuance. However, issuance of IPOs does not reduce the extent of underpricing of the followed up IPOs. We further show that IPO initial return is not related to the initial return of the followed up ones. The issuance of IPOs cannot be attributed to the information of initial returns of preceding IPOs. Rather, the market information between IPO filing date and IPO issuance date is the cause for the lead-lag relation between IPO initial return and IPO issuance. IPO offer price willfully reflect to the negative recent market return but simply partially to the recent positive market return. Most of the IPO initial return can be explained by the information revealed after offer price has been set implying that the offer price is set efficiently. The large amount of IPO volume following high IPO initial return can be attributed to the positive market reaction to the preceding IPOs instead of the filed IPO pricing of preceding IPOs. Our findings explain the hot-issue anomaly. During hot markets, investors' excess demand on IPOs leads to high initial returns of IPOs. Faced with investors' excess demand, issuers attempt to issue IPOs to take advantage of investors' sentiments to maximize IPO proceeds.